PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requeSted.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 K:/Proj/Aoc&Pres/CIRCIDaIeDue.indd

BATTLES AS INFORMATION: DOMESTIC OBSERVERS, THE EXECUTIVE,
AND COST-BENEFIT ASSESSMENTS DURING WAR.

By

Kristopher Barrett Grady

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Political Science

2008

ABSTRACT

BATTLES AS INFORMATION: DOMESTIC OBSERVERS, THE EXECUTIVE,
AND COST-BENEFIT ASSESSMENTS DURING WAR.

By

Kristopher Barrett Grady
The received literature indicates that domestic political institutions afford accountability
over the executive of a state. This finding is especially applicable to war policy given the
enormous costs and stakes endogenous to a war. The constraints placed upon the
executive are exerted by a group of actors who have the ability to remove and/or punish if
in a post-hoe review a war turns out to be sufficiently deleterious to their interests. The
specter of future sanction is thought to condition executive behavior, but what if domestic
political figures do not forestall judgment until the end of a conflict? Assuming the
ability of the enfranchised members of a state to evaluate ongoing war policies, evidence
of domestic political pressure might inspire a change in these policies prior to the end of a
war. Instead of war outcome and duration being determined by the prospect of sanction
after a war’s conclusion, the threat of executive sanction might affect the conduct of the

war well before its conclusion.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. v

LIST OF FIGURES ................................................................................ vi

INTRODUCTION ..................................................................................... 1

CHAPTER 1

THEORY ............................................................................................. 1
Literature Review, .......................................................................... 1
Sources of Domestic Political Pressure ................................................... 7
Reducing Domestic Political Pressure: Repression .................................. l8
Reducing Domestic Political Pressure: Bargaining Position ...................... l9
Reducing Domestic Political Pressure: Strategy ...................................... 21
Summary ................................................................................... 23

CHAPTER 2

RESEARCH DESIGN ........................................................................... 25
Case Selection and Data Collection .................................................. 28
Plan of Dissertation ...................................................................... 31

CHAPTER 3

AUTOCRATIC CASE STUDIES .............................................................. 34
Japan: World War II ..................................................................... 39
People’s Republic of China: Korean War ............................................. 56
Iran: Iran-Iraq War ....................................................................... 71
North Vietnam: Vietnam War ........................................................... 85
Germany: World War II ................................................................ 100
Iraq: Iran-Iraq War ...................................................................... 112
Soviet Union: World War II .......................................................... 122
Conclusion .............................................................................. 130

CHAPTER 4

DEMOCRATIC CASE STUDIES ............................................................. 139
United States: Vietnam War .......................................................... 141
United States: Korean War ............................................................ 161
United States: World War II .......................................................... 174
Conclusion ............................................................................... 1 86

CHAPTER 5

CHANGES IN NEGOTIATION POSITION ................................................ 188
Issue-Goods Negotiation Position .................................................... 190
Political Pressure ....................................................................... 194
Battles .................................................................................... 200
Hypotheses Tests ....................................................................... 203

iii

Analysis ................................................................................. 207

Conclusion ............................................................................... 228
CHAPTER 6
CHANGES IN STRATEGY .................................................................... 232
Strategy Change .......................................................................... 233
Hypotheses Tests ........................................................................ 237
Analysis .................................................................................. 239
Conclusion ............................................................................... 246
CHAPTER 7
SUMMARY ...................................................................................... 251
Findings ................................................................................. 25 1
Theoretical Contributions ............................................................. 255
Insights for Public Policy ............................................................. 259
Future Research Project ................................................................ 260
REFERENCES .................................................................................. 263
iv

LIST OF TABLES

Table 1: Cases and Number of Observations ............................................
Table 2: Frequency of Negotiation Offers ................................................
Table 3: Rare Events Logit Initiators Only ..............................................
Table 4: Rare Events Logit Sub-Status Quo Offers All Cases .......................
Table 5: Multinomial Logit Autocratic Cases ...........................................
Table 6: Rare Events Logit Status Quo Offers ..........................................
Table 7: Rare Events Logit Category Sub-Status Quo Offers Autocratic Cases. .
Table 8: Change in Predicted Probability of a Category Sub-Status Quo Offer.. ..
Table 9: Rare Events Logit Status Quo Offers Democratic Cases... .. . ... . . . .. ..

Table 10: Change in Predicted Probability of Status Quo Offers ....................

Table 11: Definitions of Change in Negotiation Offers... ... ... .... ..

Table 12: Rare Event Logit Upward/Downward Movement ..........................
Table 13: Change in Predicted Probability of Movement in Negotiations ..........
Table 14: Rare Events Logit Attempted Strategy Change Democratic Cases... . . . ..
Table 15: Predicted Probability of an Attempted Strategy Change ..................
Table 16: Rare Events Logit Attempted Strategy Change Autocratic Cases .......

Table 17: Changes in Predicted Probability of Attempt to Change Strategy. . . . . .

..... 31

...l94

....210

..... 211

....213

....216

.217

...219

.222

....223

.. .....225

....226

....227

.241

...242

....243

.244

 

LIST OF FIGURES

Figure 1: Fighting and Support for Remobilization .......................................... 9
Figure 2: Distribution of Elite Opinion in Japan ............................................ 49
Figure 3: Distribution of Elite Opinion in China ................................ ’ ............. 66
Figure 4: Distribution of Elite Opinion in Iran .............................................. 75
Figure 5: Distribution of Elite Opinion in North Vietnam ................................. 93
Figure 6: Distribution of Elite Opinion in Germany ....................................... 107
Figure 7: Distribution of Elite Opinion in Iraq ..................... . ........................ 1 16
Figure 8: Lowess Graph of Public Support for Vietnam War and War Duration ...... 144
Figure 9: Percentage of Public Supporting Escalation and Withdraw .................. 146
Figure 10: Lowess Graph of Public Support for Korean War and

War Duration .................................................................................... 164
Figure 11: Percentage Supporting Bombing China

and Assisting Nationalist Invasion of Mainland ............................................. 168
Figure 12: Lowess Graph of War Duration

and Monthly Battle Deaths ..................................................................... 170
Figure 13: Lowess Graph of War Duration

and Approval of World War II ............................................................... 177
Figure 14: Lowess Graph of Anticipated War Duration

and Win Ratio for European and Pacific Theatre .......................................... 180

Figure 15: World War II US Monthly Battle Deaths .................................... 182

vi

 

 

Chapter One: Theory

Introduction

The received literature indicates that domestic political institutions afford
accountability over the executive of a state. This finding is especially applicable to war
policy given the enormous costs and stakes endogenous to war. The constraints placed
upon the executive are exerted by a group of actors who have the ability to remove and/or
punish if in a post-hoe review a war turns out to be sufficiently deleterious to their
interests. The specter of this future sanction is thought to condition executive behavior,
but what if domestic political figures do not forestall judgment until the end of a conflict?
Assuming the ability of the enfranchised members of a state to evaluate ongoing war
policies, evidence of domestic political pressure might inspire a change in these policies
prior to the end of a war. Instead of war outcome and duration being determined by the
prospect of sanction after a war’s conclusion, the threat of executive sanction might affect

the conduct of the war well before its conclusion.

Literature Review

While not all crises end as wars (Snyder and Diesing 1977), the issue-goods at
stake in a dispute define the benefits of fighting.l The failure to satisfactorily settle upon
the distribution of the issue-goods results in the outbreak of hostilities. A state’s decision

to engage in a war, or provoke a crisis, is theorized to be contingent upon a cost-benefit

 

' Diehl supports the utilization of issues as a key component in understanding the origins of international
crises and subsequently wars (1992). The most frequently cited issues over which a crisis emerges are
categorized by Holsti as tenitory, national liberation/state creation, commerce, enforcement of treaties and
state/regime survival (1991). Gochman and Leng’s survey of issues finds that the most salient in terms of
sparking militarized disputes is that of territory (1983).

l

calculation made by the initiating state (Bueno de Mesquita 1980). A favorable balance
in the cost-benefit calculation by a leader for initiating a conflict and a defending leader’s
similar assessment to resist rather than accept a negotiated settlement, results in war.

War onset has been represented as the failure of opposing states to find an
agreement that divides the issue-goods in question (F earon 1995). Acknowledging that
war is ex-post inefficient, why then would states engage in a war? The indivisibility of
issue-goods, inability to guarantee agreements, and misrepresentation of private
information are among answers offered for bargaining failure. F earon’s model depicts
the onset of a conflict as a suspension of the bargaining process until the war has
determined the victor and the vanquished with the attendant distribution of issue-goods.
But if a war is projected to be superior to other policy alternatives prior to the outbreak of
a war, then why do wars end?

The war termination literature has expanded upon these results in pursuit of the
causal mechanics compelling a negotiated settlement. The empirical puzzle that these
pieces seek to address is that unlike Fearon’s parsimonious model (1995), most wars do
not conclude with a disarmed opponent surrendering more or less unconditionally.2 More
typical is what Clausewitz called a “real war” resulting in some intermediary settlement
(Wagner 2000, 479). As the actual ability of states to obtain the issue-goods being fought
over in a conflict is revealed, an agreement is reached as to the likely outcome of a war
prosecuted to the end. But what kind of information is transmitted through the course of
a war that enables a mutual recognition of the likely outcome?

Wittman contends that hostilities conclude after both parties view peace as more

favorable than continued fighting in a cost-benefit analysis (1979). The expected utility

 

2 Effectively defeated states are thought to make the initial overtures for peace (Beer and Mayer 1986).

2

of fighting changes throughout the course of a war as previous expectations of costs and
benefits are altered. As battles indicate which side is militarily superior, a peace
settlement is made that divides issue-goods benefits on the basis of the mutually
understood distribution of capabilities (Blainey 1988).3 In effect bargaining is not
suspended after the onset of hostilities, but continues until a mutually acceptable
distribution of issue-goods is determined. The ongoing bargaining model of war
conceptualizes how fighting helps to facilitate a settlement even if no actual negotiations
are taking place.

The costly, therefore credible, information created through fighting is attributed
with the ability to reveal a state’s capabilities and/or resolve (Wagner 2000). The
revelation of private information facilitates the conclusion of a conflict, a phenomenon
known as the “principle of convergence”. But if states know themselves to be
comparatively weak or unresolved then why do they willingly participate in a war?

Filson and Werner theorize that states update their beliefs with respect to the
relative capabilities of the states fighting (2002). Because states have an incentive to
misrepresent their true capabilities and resolve in order to attain a better negotiating
position. States that posture themselves as being stronger or more resolved than in fact
they are can potentially be defeated militarily after a war’s onset.4 Military collapse by

one of the combatants, especially after a comparatively small number of battles, is not

 

3 Smith illustrates that battles dictate the eventual outcome of a war, as opposed to a fixed set of capabilities
assessed at the beginning (1998). The incremental nature of war bargaining, disaggregating wars into their
component parts is found to be instructive for determining the eventual outcome.

’ Powel makes a similar argument based on a formal model that the attacking state both fights and makes
negotiation offers in an attempt to decipher the defensive state’s “type”, being either strong or weak (2004).

3

especially frequent in the historical record limiting the empirical applicability of the
theory.5

Slantchev offers a somewhat broader model (20033). Generally a war ceases
because of a mutual recognition of the relative strength and resolve of the opponents
becomes clear through battles, with the commensurate distribution of issue-goods being
doled out with a settlement.6 The informational pr0perties of fighting are further detailed
by the finding that relatively weak states are more adroit in the accumulation and
acceptance of new information within a conflict (Slantchev 2004). While states are
differentiated by relative power, an equally important unit-level distinction is regime
' type. i

The leader of a state is beholden to a certain constituency, “selectorate”, the size
of which is contingent upon political institutions. In a democratic context the selectorate
is the comparatively large enfranchised portion of the population. Alternatively, within
autocracies this group is defined as the coalition of actors that maintain a leader’s
position or members of potential coalitions that could deselect a leader from power
(Bueno de Mesquita et al. 2003). Upon participating in a war the outcome of the conflict
is the criteria by which the executive’s policy choice is assessed.

Even a minor war defeat likely results in the loss of power by democratic leaders.
Whereas autocratic regimes can repress the opposition and retain power in all but the

most catastrophic of losses (Bueno de Mesquita et al. 2003). Goemans theorizes that

 

5 Many comparatively weak states in asymmetrical conflicts do better than anticipated (Maoz 1989).

6 Another theoretical narrative of the informative battles theory is that even with complete information,
regarding the actual balance of resolve and capabilities, states may continue to battle in a seemingly
inefficient manner (Slantchev 2003b). Despite the fact that a state knows it will ultimately lose, the weaker
state continues to fight and inflict costs in order to gain a stronger negotiating position for the termination
of hostilities.

4

 

both democratic and autocratic regimes are only likely to face punishment, as well as
removal from power, after disastrous defeats for their country (2000). By contrast, semi-
autocratic states are likely to face punishment for even moderate defeats, due to their
more limited ability to suppress opposition. Consequently these states are theorized to be
risk acceptant in how they prosecute a war.

Regardless the sanctioning mechanism of domestic interests is what conditions
the choice of what wars are fought. The executive’s constituents are theorized to defer to
the executive with respect to war mobilization confident that the leader is accountable on
the basis of a post-hoe assessment of a war’s utility. Why then does the selectorate
acquiesce to' mobilization prior to a war’s onset if they can only deselect the executive
after a post-hoe review?

Gaubatz demonstrates that while the executive in a democracy is not necessarily
obstructed from undertaking belligerent actions, elections do offer opportunities for
political opponents to articulate policy alternatives (2000). However, in even the most
liberal democracies the government dominates national security information. The
executive’s ability to select wars is combined with state domination of security
information creating the conditions for the instrumental use of force (Ostrom and Job
1986). But if there is an anticipation of limited public goods utility, then why does the
democratic public mobilize in the first place?

Institutions within a liberal regime help diffuse information to the broader public
after the fact (Colaresi 2007). Stronger institutions providing outlets for public review of
national security information reduce the likelihood that force will be used for ulterior

motives, consequently bolstering support for mobilization. The greater the level of public

support for war mobilization, the more likely that a democracy will be successful in
militarized disputes. In democracies a post-hoe review of war policies likely results in
electoral sanction if a war entails high costs or is fought over inappropriate objectives
(Downs and Rockel992). What defines these costs that reduce the public’s perceptions
of a war’s utility?

Referencing casualties as costs, it is found that democracies are casualty averse,
with growing numbers of human losses diminishing public support for a war (Gartner and
Segura 1998; Mueller 1973). A democratic public is theorized to assess casualties and
weight these costs against the benefits of fighting. Metrics such as body counts,
battlefield victories, and the importance of the war help define success (Larson 1996;
Feaver and Gelpi 2004). As both domestic actors and the executive are privy to similar
updates in terms of costs and benefits it appears that assessments of a war’s utility vary
between the public and the executive. As a war lengthens and the toll in lives grows,
democracies have a greater difficulty mobilizing the populace for war and subsequently
are less likely to win over time (Reiter and Stam 2002).

The wide range of interests represented in a democracy potentially weight costs
differently than does the executive. However, does the smaller number of enfranchised
actors in an autocracy mean that assessments of costs and benefits are comparatively
homogenous?

Peceny et al. illustrate some of the distinctions between autocratic states as
different political norms and structures condition policy outcomes (2002). Single party
states are more constrained due to the higher likelihood of a leader being replaced when

compared with other types of autocracies. Single party states also feature a greater

transparency in policy decisions than do military juntas or personalistic dictatorships.
These distinctions across autocratic regimes illustrate the potential for policy differences
between the effectively enfranchised elites and the executive.

The costs incurred by a state during a war include casualties, financial costs, loss
of international prestige, loss of the ability to repress dissent, and a state’s physical
destruction (Pillar 1983; Goemans 2000). The elites in an autocracy are shown to
sometimes assess the cost differently than does the executive of the state. Mendelson
notes that foreign policy changes can occur in autocracies when an epistemic community
gains a hearing from relevant political leaders (1993). A review of policies is conducted
by elites prior to a war’s conclusion.

Because the preferences of these two groups can differ substantially a retraction in
support for mobilization is possible. As with the democratic casualty literature it appears
that autocratic elites can assess a war’s net utility prior to its conclusion. Especially if
losing a war might mean that their ability to sanction the executive after the fact is
compromised due to war outcome.7 Why would a democratic public or autocratic elites
acquiesce to further mobilization if they anticipate a negative outcome to the war? If
discontent is emerging with war policies would the executive not try to mitigate political

pressure through an alteration in war policies rather than face the prospect of sanction?

Sources of Domestic Political Pressure
Two major problems emerge when surveying the received literature. First, there
is an apparent disconnect from the political institutions literature (Bueno de Mesquita et

al. 2003; Goemans 2000) and the possibility of domestic intra-war review of war policies

 

7 This could occur due to the regime’s collapse in revolution or destruction by foreign conquest.

7

(Gartner and Segura 1998; Mendelson 1993). The expectation that the executive
anticipates and fears sanction and selects war policies accordingly, is a cogent theory.
But is it a realistic expectation that domestic actors will not assess the relative merits of a
war on the basis of the current policies, rather than suspend an evaluation of a war to the
very end?

The second major problem is nested within the first. If intra-war updates as to
costs and benefits is possible, this implies an ability by domestic actors to assess costs
and benefits in a way similar to that of the executive. The principal of convergence
literature tacitly assumes a unitary actor, while the more empirically oriented literature
suggests that this is likely an important oversight (Mendelson 1993; Gartner and Segura
1998; Feaver and Gelpi 2004). If assessment of war policies is conducted prior to the end
of a war, how do domestic political actors respond to a diminished appraisal of a war’s
utility?

I contend that the mobilization, post-hoe war appraisal, sanction model might not
operate in a single cycle. Instead in a longer war a series of remobilizations are likely
necessary to prosecute a conflict, with the relevant domestic political actors equipped
with the ability to register discontent with war policies prior to the conclusion. I theorize
that these domestic political actors are those members of a polity that have the ability to
sanction through deselection or punishment the executive of a state (Bueno de Mesquita
et al. 2003; Goemans 2000).

As a point of clarification, two different dynamics are present in terms of a cost-
benefit appraisal of a war. Initial war mobilization is theorized to be conditioned upon

domestic political actors and the executive jointly assessing a war as a net positive in

terms of expected utility (Bueno de Mesquita 1980). In as much as wars are typically a
negative sum game, most states experience a net loss by the end of a war (Fearon 1995).
It is likely that many domestic political actors will therefore assess a net utility loss for a
substantial portion of an ongoing war. The divergence in a conflict’s assessed utility is
contingent upon if plausible alternatives to current war policy would enhance domestic

political observers’ assessments of a war.
Figure 1: Fighting and Support for Remobilization
Failure of Negotiations

1

Domestic Mobilization 4. . . . . .

l 5.3 Updated Beliefs of a
‘.. .. c 0
.' War’s Net Utlllty

Executive Decision Making .3

1

Fighting
Benefits = (probability of success X utility of security goods) / time discount

Costs= (casualties + other costs)

Given current war policies and what has been learned through fighting would an
alternative set of policies, different from those being pursued by the executive, enhance a
domestic political actor’s utility. To the extent which the executive pursues policies that

conform to their unitary preferences that simultaneously do not maximize the expected

utility of domestic political actors a rift in policy preferences occurs. Extrapolating into
the future domestic political actors that perceive current war policies as sufficiently
harmful to their interests might refuse to support the executive and/or refuse to mobilize.

The public has the ability to observe and make cost benefit assessments on the
basis of the information garnered through fighting. The relative transparency by which
the public assess a war is afforded by such devices as a free press, legislative oversight,
and freedom of information laws (Colaresi 2007).

Domestic political actors within democracies can deselect an executive from
power via elections. In the intra-election or lame duck period a loss of political clout to
pass legislation or other domestic priorities might be another deterrent.8 Reputational
costs to a leader’s party and in extreme cases impeachment might also serve to enforce
the interests of the executive’s principals (Cotton 1986; Segura and Barratt 2004). Short
of deselection a refusal of the population to support further mobilization is also possible.
A prominent example would be the American public’s dissatisfaction with war policies
leading to the refusal of the Congress to continue funding the war in Vietnam.

In the autocratic context a smaller cohort of individuals compose the class of
domestic political actors. These elites are considered to be enfranchised in the sense that
they can serve as potential supporters in forming a new coalition that selects an
executive. In addition to their effective enfranchisement these actors also have the ability
to observe battlefield events due to their positions of influence.

To further differentiate the domestic political actor from the general public in the

autocratic cases, these actors’ ability to deselect the executive is distinct from a broad-

 

’ Presidential approval are at least partially salient in explaining the likelihood of legislative success in
Congress (Canes-Wrone and de Marchi 2002).

10

based revolution. Meaning that instead of pursuing a wholesale change of the current
governmental structure, the domestic political actor has a position of import in the current
regime. An example would be the political infighting in the Hanoi during the Vietnam
War. None of the leading members of the Politburo sought to change the form of
government instead they sought to alter the executive. To illustrate this class of actors a
couple of examples are highlighted.

In the Soviet Union during World War II the center of power was in the Politburo
under the dominance of Joseph Stalin. The relevant domestic political actors consisted of
those general officers in the Red Army, the NKVD (People's Commissariat for State
Security), or other political-military organizations. These Soviet citizens had access to
the information relevant to forming a costs and benefit analysis of a war, but also held
positions of influence that when aggregated conceivably would be able to deselect the
executive.

The Third Reich’s domestic political actors consisted of individuals that held top
positions in the Nazi Party and/or within the government as well as high-ranking military
officers or internal security forces such as the Gestapo and the SS (protective squadron).
These positions allowed for both the potential to supplant the executive and assess the
costs and benefits of fighting in spite of state domination of the media and internal
repression of the broader public.

In both the autocratic and democratic context the potential for discontent and an
intra-war review of executive policies is theorized as possible. However if an executive
is selected by domestic political actors, then why would their interests substantially

diverge over time? Executive selection or re-selection should give domestic political

ll

actors leaders somewhat in line with interests (F earon 1998a). If the assessments of cost
and benefits diverge between the executive and domestic political actors, two major
factors are theorized to be at the center of these divergent impressions of a war’s utility.

First, the information that is received by the domestic political actors is
potentially less voluminous and less accurate than what is received by the executive. The
executive may have greater access to information than do domestic actors due to
intelligence, military reports or other sources that are, at least for a time, kept secret. This
is a plausible argument in the democratic context, where information regarding military
secrets is by necessity guarded from the public. However, information does become
public and costs and benefits can be assessed even if there is a lag between when
battlefield events occur and when the public receives the information (Colaresi 2007).
Examples might include the Pentagon Papers, the outcome of battles such as D-Day or
Tet Offensive, and/or reports of battlefield fatalities.

Within the autocratic context a substantial differentiation in received information
is somewhat less plausible. According to the conception of domestic political actors
offered above, the effectively enfranchised members in these societies hold positions of
influence. These members of the upper echelons of the military, intelligence/domestic
repression apparatus, bureaucracy, and party officials have information that is otherwise
denied to the general public. In fact given their positions of influence, some domestic
political actors might be more adequately informed as to a war’s developments than the
executive.

The alternative explanation for different assessments of battlefield events is that

the executive and domestic political actors have fundamental differences in how they

12

assess costs and benefits. The differences in these assessments might be found in how
different types of costs are weighted, the assessed utility of the issue-goods being fought
over, and discordant impressions of the probability of achieving these benefits or the time

until benefits are realized.

benefits @robability of success X utility of security goods) / time discount > costs (casualties + other costs)

A potential source of differentiated policy analysis might be the interpretation of
costs. It is a widely accepted belief that democratic publics are highly averse to
casualties (Gartner and Segura 1998; Mueller 1973), with the public frequently retracting
support as the number of battle deaths rises. The duration of a war is also thought to
undercut support as costs are incurred, reducing a democratic public’s willingness to
mobilize (Reiter and Stam 2002). Other costs of a war (lives, resources, opportunity
costs) might also be weighted differently by domestic political actors as opposed to the
executive.

An additional factor to be considered is that the costs of a conflict accrue
comparatively quickly, whereas benefits are usually not realized at the same rate.9 If the
time discount that the domestic political actors apply to achieve the issue-goods is
substantially different from that which the leadership of a state employs then in spite of

comparatively low costs the public may balk at continued mobilization.lo

 

9 The disconnect between the rate at which costs and benefits are realized is an important question in
addressing the public’s support for a military intervention, the historical reference of a state might serve to
illuminate the benefits of undertaking a conflict (Pickering 2002).

'0 This sentiment is most famously articulated by General George Marshall’s statement that a democracy
cannot win a seven-year war (Reiter and Stam 2002, 164).

13

Among the most plausible reasons for divergence in the assessed utility of a
conflict is if the issue-goods stated in the case for mobilization are not the same ones that
a state pursues after a war’s onset. While domestic political actors collectively
acquiesced to military mobilization, a war’s scope and objectives might have expanded
beyond the initial purpose or inspired balancing reaction from other states (Snyder 1992;
Walt 1987).

A further source of differentiated interpretations of costs is that in the process of
garnering support for war estimations for the cost of a conflict are sometimes made.
However, such estimates are usually considerably lower than actualized costs (Nordhaus
2002). While these assessments are occasionally made publicly in a democratic context a
similar sales pitch could also take place in an autocracy (Snyder 1992; Goemans 2000).
Regardless, these statements might introduce audience costs in which estimations of
expected utility are declared and reviewed later on (Fearon 1994). If actualized costs are
incongruent with previous estimates this may exacerbate negative assessments of the
executive’s war policies.

The case for war might also have been made for misleading or diversionary
purposes, not identified when the initial case for mobilization was laid out (Ostrom and
Job 1986; Russet 1990). As a result the domestic political actors might have a reduced
sense of utility in the continued prosecution of a conflict.11

In an autocratic context support amongst domestic political actors for diversionary

war might be stronger, for example new regimes might use wars as a means to

 

" The empirical foundations of the diversionary war hypothesis are not as strong as some would suggest
(Levy 1996).

14

consolidate their position.12 While a diversionary war might benefit the executive of a
state, autocratic domestic actors within these regimes might compute the risks of a
conflict to be less valuable than diversionary gains. The executive might be likely the
primary beneficiary, effectively reducing the domestic political actors’ relative position.

The domestic actors in an autocracy may interpret a heightened risk of revolution
in the state if a war becomes especially costly. The increased burdens that a war places
upon the citizenry and the potential degradation of the repressive apparatus of a state due
to combat could heighten the perceived risk of revolution.13 A change in regime would
carry heavy costs, even if members of the elite should escape punishment, the private
goods derived from autocratic domestic actors’ privileged position is potentially a major
expense. Short of revolution, the elites of an autocratic state might resent the diversion of
public resources to a conflict and away from more tangible private goods.

If domestic political actors observe fighting and observe a reduction in utility on
the basis of some objective criterion like casualties or financial expense, why do
executives pursue policies at odds with these assessments? An executive can become
politically invested in a war over which he/she serves as leader such as Churchill or
Hitler. The imperative for an executive to successfully prosecute a conflict to which they
are closely identified might cause them to stress sunk costs at the expense of future
debits. In effect the executive of a state is likely to judge the utility of the issue-goods at
stake in a conflict more favorably relative to costs than do domestic political actors.

The executive might also fashion a previous psychological impression that more

readily allows for the assimilation of positive rather than negative information updates

 

’2 Both post-revolution Iran and China serve as examples.
‘3 The Russian Revolution serves as a prominent example of this possibility.

15

(Khong 1992). A leader might also perceive a greater likelihood of achieving the issue-
goods sought after in a war when compared with domestic political actors.

I do not contend that all domestic actors are uniform in their analysis of a conflict.
However, it is the assumed that a significant portion of a leader’s selecting domestic
actors may withdraw their support given different expectations of a war’s utility. Factors
such as partisanship, ideological orientation, role in a government, and bureaucratic
politics can all influence how an executive’s constituents assess policies (Allison 1971;
Gelpi and F eaver 2004). The heterogeneity between enfranchised domestic political
actors can lead to the application of political pressure as some of these observers assess
war policies less favorably than others. If alternative war policies appear to have greater
utility than those policies currently pursued by the executive this can result in the
application of political pressure.

Political pressure is exerted upon the executive through a few different outlets.
Short of executive deselection, in a democratic context these include refusal to support
mobilization or other policies of the state’s leader. It might also include sanctioning
other co-partisans at the ballot box and demonstrations or other acts of civil unrest.

In an autocratic context domestic political actors have a somewhat different set of
options in redressing their concerns over war policy. '4 One outlet is lobbying the
executive to alter war policies. Autocrats enjoy positions of influence and are well-
placed to have their opinions heard by the executive. This avenue does have the risk of
inviting repression by the executive, but it is fairly common for civilian and military

elites to articulate alternative policies.

 

" Sometimes military officials will circumvent executive orders in order to execute a military strategy
more to their preferences, although usually not on the scale necessary to fundamentally alter the course of a
war, an occasional occurrence in Nazi Germay (Irving 1977).

16

f?

 

The principal consequence of dramatically different cost-benefit calculations, is
the deselection and/or punishment of the executive. In the democratic context a majority
or plurality of voters determines elects the executive of a state at the institutionally
mandated time.

By contrast deselection of an autocratic leader is a very hazardous pursuit. The
latent sentiment of these domestic political actors have for their leader these opinions are
largely unregistered in any public forum. Only those participating in an explicit effort to
remove the executive are known to be “voting” for a new leader. This makes identifying
the size of a credible deselecting coalition difficult.

Serious attempts at a coup or assassination would like reflect the sentiments of a
majority or some substantial plurality of domestic political actors. Noting the hazard of
attempts to apply political pressure upon the executive the collective action problem is an
issue, suggesting that those actors participating in an attempt at deselection are
representative of a larger body of actors of a similar disposition (Olson 1982). To the
extent that domestic political actors would hazard such an attempt the current inertia of
war policies must constitute a major threat to their perceived utility.

Given the stress that the literature has placed upon the importance of sanction and
punishment for a leader, an executive approaching this level of political pressure would
try to accommodate the grievances of domestic political actors. If burgeoning political
pressure can be ameliorated, what options does an executive have in order to mitigate the

possibility of deselection?

l7

4' ’4'
thusll

new?

H“
can;

N313”
v “L‘A-
l

'33 ‘ll

Tl"
‘Cllli‘

V
293*
‘HV\

net

Reducing Domestic Political Pressure: Repression

Through the course of the above discussion it has generally been assumed that the
executive of a state has a higher net utility assessment of a war than their associated
principals. It is possible that the opposite is true, although this appears to be infrequent in
the cases evaluated below.” There are several empirically observable phenomena that
potentially might act as mechanism for alleviating political pressure. The most obvious is
the utilization of repression for the purposes of coercing compliance with the current
military policy.

For the purposes of my dissertation it is assumed that only autocracies have an
effective repressive apparatus. Despite obvious anecdotes such as the internment of
Japanese Americans, McCarthyism, and the banning of radical political parties, the
ability to systematically repress is largely absent in democracies. While autocracies are
not homogenous, these illiberal regimes all possess at least some capability of quashing
public dissent. This is accomplished by a litany of nefarious mechanisms secret policy,
martial law, arbitrary imprisonment and the like. The principal distinction that I would
draw is the repression of military or political elites within a regime, rather than the typical
coercion meted out to the broader public. While certainly not unheard of for an
autocratic state to purge party members, military officials or even the executive in

peacetime, these behaviors are assumed to be exacerbated when an autocratic state is at

war.

 

'5 If these opinions constituted a major portion of the domestic political actors, it would be expected that
greater exertions should be undertaken in order to secure the benefits at stake. While there is anecdotal
evidence for this, public hawkishness with respect to Vietnam and Korea and Japanese military leaders in
World War II, on balance political pressure for greater efforts in fighting a war appears to be much less
fi'equent than the opposite sentiment.

18

Re

mm o

1‘

01 inlet .

. SilllC. l
V

_‘I .
litdliillllj

\W 8le

Ill

(’1
CA

0;,3.
waxy

,(3

Repression of elites is taken as a means by which an executive can prevent an
erosion of his/her political standing or possible removal. Potential rivals for the position
of chief executive would likely emerge from the higher echelons of power in the current
regime, excluding a broadly based revolution. As such most repressive effort should be
directed against elite officials in both the government and military as the likely
candidates to form an alternative coalition. Yet not all autocracies have a developed
capacity to repress elites and must rely on alternative means of relieving political

pressure.

Reducing Domestic Political Pressure: Bargaining Position

Repression can be inefficient, expunging military and/or bureaucratic expertise,
lowering morale amongst elites, reducing policy innovation and autonomy. Additionally,
it is difficult to identify the actual dissenters exacerbating these inefficiencies. Given
these expenses what policy changes can be undertaken in order to arrest political
pressure?

As noted above, the issue-goods at stake in a war are not necessarily static or uni-
dimensional (Goemans 2000; Mclnnis 1986). The ongoing bargaining model of war also
highlights the progressive movement to a mutually acceptable conclusion of hostilities
(Slantchev 2003a, Wagner 2000). The essential tradeoff is the value of the issue-goods
bargaining position that a state establishes with the commensurate likelihood that this

position will be accepted by an opponent (Pillar 1983).16 I contend that an alteration in

 

'6 The willingness to negotiate itself might serve as an important factor in a peace settlement (Pillar 1983).
If disingenuous bargains have costs in terms of future credibility (Fearon 1998) then this could serve as a
credible signal to domestic political actors that a settlement is comparatively imminent.

l9

fit‘llil

3583511

midi h
objecti‘
series
upper:

30.720

4;“

axOl
MAS"
Older
e _ ll
l1U'ul

K T‘
1 “-Ir
W All

negotiation position might have the effect of altering the domestic political actors’
assessments of the net utility of a war.

Domestic political actors might be convinced that a settlement of lesser utility
might have a greater overall value in that the perceived probability of achieving the new
objective in a war is substantially higher. Reducing sought after benefits, thereby
increases an opponent’s net utility increases the probability of acceptance by an
opponent. A more conciliatory negotiation offer also potentially shortens the time
horizon of the length of a war and lowering domestic political actors’ time discount upon
the realization of benefits. As illustrated by Hanoi Politburo members heavily
discounting the value of reunification based upon the time until it was likely to be
achieved. Or elites in the Chinese Communist Party wanting to begin negotiations in
order reduce the expectation of future costs of fighting in Korea.

Rather than accepting reduced terms of a settlement, an alternative policy option
would be to attempt to enhance the mobilization of a state in order to better prosecute the
conflict. The obvious drawback from this policy option is that it would increase costs as
well, with the hope of enhancing the perceived probability of achieving the issue-goods at
stake in the war. An executive already facing discontent, might find maintaining the
current level of mobilization into the future is likely to be a difficult enough
proposition.l7 An exception to this assertion would be that heightening resources devoted
to a war might be viewed as a superior policy by domestic political observes to the

prospect of defeat, particularly in a highly salient conflict.

 

'7 An executive might also attempt to mobilize further and then increase its bargaining position in terms of
issue-goods enhancing the utility of a conflict (Goemans 2000). Given my underlying theory that domestic
political pressure is undermining a leader’s position increasing mobilization for heightened benefits given
the already disappointed returns that the domestic political actors already observed.

20

Reducing Domestic Political Pressure: Strategy

An alternative policy option for an executive is to alter the perceived efficiency of
how a state undertakes a conflict. Strategy alteration might enhance domestic political
actors’ assessed utility by lowering casualties, reducing material losses, improving the
prospects for victory or hastening the end of a conflict. A superior strategy may be able
to take the same resources devoted to a conflict and either enhance the overall perceived
benefits of a war or reduce costs.

A discussion of strategy in war is only relevant to altering the level of domestic
political pressure if the executive has some meaningfirl control over strategy choice.
Mearsheimer contends that political leadership helps dictate the form that strategy takes,
referencing Hitler’s influence on Germany’s choice of a blitzkrieg strategy (1999).
Reiter and Meek note the influence of regime type as being an important determinant of
what strategy choice with democracies using the relatively quick and less costly
maneuver strategies (1999). The American leadership’s choice to fight limited wars in
both Korea and Vietnam compelled the selection of an attrition strategy to meet their
objectives (Gartner and Myers 1995). Given the seeming importance that the executive
has in determining strategy, what impact do these tactics have on military outcomes?

Strategy has been shown to be an influential determinant in terms of both the
success and duration of wars (Stam 1996). Military tactics have also been demonstrated
to be a determining factor regarding the outcome of battles and wars, more important
than numerical superiority, material capabilities, and weapons technology (Biddle 2004).

Studying the United States and Britain, Gartner demonstrates how key metrics are

assessed to define the success of a given strategy. When these metrics illustrate a

21

A.)
‘

1!?
E

my .3“

.4» )i
l

‘1
II‘ .

Minnie

I ,1
‘l'S-u
“ab“:

4.
I)

\-a

bl

q“.

r _ n-
AT 1

("I

1411

H‘

disconcerting trend this provides the impetus for a change in tactics (1997). Strategies
appear to have a bureaucratic inertia that is arrested only under duress. The costs .
imposed upon a state incurred from an inefficient strategy can also be changed by
civilians. Posen illustrates that there is a pro-offensive bias in the military’s planning, but
this pathology can be arrested if the political leadership intervenes (1984). Given these
findings why would an executive continue to pursue a relatively inefficient strategy?

Political capital invested by an executive into a strategy, when altered, might
discredit the broader war in the eyes of domestic political actors. George W. Bush’s
recalcitrance to order a troop surge was doubtless influenced by the prospect of
undermining already eroding support for the war and confidence in his management of
the Iraq conflict. The possibility of political pressure emerging and reducing the
likelihood of acquiescence to further mobilization or an executive’s political support is
potentially an important source of strategy change, in spite of an executive’s preferences.

A change in strategy could affect domestic political actors’ perceptions of the net
utility of a conflict through a more efficient use of military resources. Changes in
military strategy can alter the efficiency by which states employ military resources. The
grinding attritional stalemate on the Western Front of World War I was nearly broken by
a modest alteration in German tactics near the end of the conflict (Biddle 2004). Such
that similar amount of resources being devoted to war when used differently catalyzed a
greater level of military success, enhancing the probability of obtaining the issue-goods at
stake in a war.

The Tet Offensive was crafted by Hanoi as a final push to inspire a popular revolt

and topple the Saigon government (Duiker 1980, 63). By reducing the likely time

22

horizon until the issue-goods in a conflict could be realized the pursuit of an aggressive
military campaign potentially heighted the utility of a war through a reduced time
discount on the realization benefits.

Changes in military strategy might also be introduced to lower costs or the
expectation of costs associated with other tactics. Nixon’s attempt to salvage the Saigon
government in 1972 was accomplished by the introduction of massive amounts of
airpower rather than the reintroduction of American ground forces. The blockade of
Japan, strategic bombing, and use of the atomic bomb were executed in lieu of the more
costly strategy of an invasion of the home islands.

Military strategy, issue-goods negotiation position, and in the case of autocracies
repression, are all means by which an executive can mitigate domestic political pressure.
If an intra-war assessment of a war can be made by both the executive and domestic
political actors, an evaluation of the frequency and impetus behind changes in war policy

are the empirical implications that my dissertation will evaluate.

Summary

The purpose of the following project is to define a unified theory of how states
alter their perceptions of the likely outcome of an ongoing war. If domestic political
interests can make similar assessments the divergence between these assessments is
theorized to be a source of political pressure. An executive having sunk the political
capital into the endeavor might be reticent to alter war policies in acknowledgement of a

failure to select a war well. Should intra-war assessments of a conflict be available to

23

domestic political actors, this potentially opens up the opportunity for domestic pressure
to enforce changes upon the executive.

More broadly it has generally been democratic states that have been defined in
terms of domestic political interests substantially altering war outcomes. Casualties and
war duration detract from public support for war mobilization or democratic executives
being compelled to select well the wars in which they will join or initiate. But are
democracies fundamental in nature or are executive incentives merely different, therefore
autocrats are beholden to some coalition of domestic political interests. The purpose of
this dissertation is to define a theory of domestic politics and war outcome that spans

different regime types.

24

Chapter Two: Research Design

The importance of domestic influences upon an executive is generally restricted
to how wars are selected. In autocratic cases the threat of sanction at the end of a war
following a review of war policies is made by an agent’s principals or punishment by
foreign occupation. For democracies the potential that casualties and war duration will
manifest a retraction in support for the executive or the willingness to mobilize, are also
important considerations.

One problem for the preeminence of war selection is that a war begun with a
given set of assumptions or objectives can change radically over the course of fighting.
Whereas informational asymmetries might allow for a comparatively ready acquiescence
by domestic interests to join a war, these interests might not suspend reassessment of a
war until it has concluded. Further wars can drag in more states, military objectives
change, and the balance of forces between states can dramatically shift making the utility
calculations of the outset of a war less relevant to the eventual outcome.

Implied in the war selection is that a unified appreciation of battlefield updates
compels a state to accept a reduced issue-goods settlement at the end of a conflict. An
executive that pursued a war that ends badly is compelled to come to grips with this.
However if there is the anticipation of punishment than more risky strategies will be
endorsed. But what if the constituents of a state do not suspend assessments until the end
of a conflict?

War selection is held to be conditioned in part by an executive’s fear of post-hoe

sanction should a conflict becomes overly costly or if a state loses catastrophically. More

25

broadly wars are initiated when both states believe that contesting an issue-militarily is a
superior option to a negotiated settlement, the selection of which wars to fight conditions
the outcome of a conflict. However if wars are reviewed prior to their conclusion, then
the selection effect of what conflicts to fight would not be as empirically salient as earlier
thought. Instead reassessment of war policies might compel the executive to change the
conduct of a war substantially altering the outcome of a war after it has been selected.

Do domestic political actors make assessments of a war’s utility previous to the
conclusion of a conflict? And do these assessments become so substantial that demands
for an alteration in war policies forms? Finally does the demand for changes in war
policies fundamentally alter how wars are conducted?

The loss of parsimony in rejecting a unitary assumption is justified only by a
substantively improved understanding of war prosecution and outcome. I contend that
before a war actually terminates, how a war is prosecuted is affected by domestic
political actors influencing the policy process. The outcome of a war may in fact be
largely determined by the input of an agent’s principals demanding changes in how the
conflict is conducted.

While domestic political pressure might result in demobilization and consequently
a reduced likelihood of success, as was the case for the United States in Vietnam. There
is also the possibility that demands for alternative war policies enhance the efficiency of a
state’s military effort as was true of the Soviet Union during World War 11. Similarly,
Khomeini was compelled to accept a reduced settlement with Iraq and Nixon was

similarly pressured to sign a reduced treaty with North Vietnam.

26

At the onset of a war, the leader’s constituents acquiesce to a war after assessing
the net utility of a prospective conflict. If subsequent impressions of war policy’s utility
relative to alternatives becomes increasingly negative, and the executives still favors
status quo policies, then political pressure may mount on the executive. As such this
divergence in opinion can threaten a leader’s political position or his ability to mobilize.
The greater a war’s duration and intensity the more clarity observers, both executive and
domestic political actors’ have for a conflict’s net utility.

War selection is thought to condition democracies to a substantial extent, driving
the finding that these more liberal states enjoy greater success in the wars they join.
Likewise when democracies are found to lose wars this is driven by casualties and war
duration decreasing a state’s willingness to mobilize. But what would be the comparable
assessment made by autocratic elites? If both are found to reassess executive war
policies previous to a conflicts conclusion a unifying explanation for policy changes in
both democracies and autocracies could be identified.

The following chapters will use the relative strengths of qualitative and
quantitative methods in order to strengthen the validity analysis. The qualitative portion
of the analysis examines how observed outcomes in the historical record come about.
Upon entering into a conflict, what drives changes in the issue-goods objectives, the level
of resources mobilized, and what military strategies are employed in the war.

If a reticent executive is found to be compelled into accepting changes in war
policies that diverge from his/her unitary preferences then this would represent evidence
against a unitary rational interpretation of battlefield events. The intra-war assessment by

domestic political actors might better define how wars are fought and concluded than can

27

be extrapolated from how domestic politics conditions selection and post-hoe review of
war policies. Additionally, the political pressure hypothesis also permits the
incorporation of strategy change and the level of mobilization as well as issue-goods
position into an analysis. All of which contribute to war outcome.

A series of states prosecuting wars are examined in order to determine if an
executive’s principals make intra-war assessments of a war’s utility. Upon establishing if
this is the case, the empirical relevance of differentiated assessments between principal

and agents is then made.

Case Selection and Data Collection

The theory presented above implies the need for remobilization of war effort,
rather than just an initial case for war that is reviewed after the conclusion of a conflict.
This winnows down the number of wars that are relevant to my analysis. The selection
of cases is informed by the need for a representative sample, adequate historical records
for qualitative analysis, and the ability to transform cases into quantitative data.
Examining the COW’s (Correlates of War) compendium of wars, defined as at least
1,000 battle deaths, most Twentieth Century interstate conflicts are comparatively short
in duration (Small and Singer 1982). Due to the brevity of most wars cease after a few

months, it is difficult to quantify political pressure and other variables.

28

The assumption that information can be disseminated to the relevant domestic
political actors is important in that recent battles are highly relevant in defining cost-
benefit assessments. This technological environment implies the wide usage of
telegraphs, radios, telephones, and other devices that permit diffusion of information
relatively quickly.18 In order to assume the rapid acquisition of information across cases,
my analysis includes World War II and conflicts occurring after 1945.

The difficulty in case selection is exacerbated by the relative paucity of military
histories for many of the COW’s identified interstate wars.19 In order to create variables
of sufficient validity within the sample requires detailed accounts, especially in terms of
defining battles and evidence of varied negotiation positions limited the size of the
sample.

Case selection was also made with respect to regime type and the relative balance
of material capabilities between the combatants. A mixture of regime types is necessary
to illustrate the importance of political pressure in both autocratic and democratic
systems. A range of autocratic regimes are included from totalitarian states like the
Soviet Union to the anacrocies of Imperial Japan and Revolutionary Iran. The different
types of autocracy indicate the relative ability of an executive to repress elites within his
state and thereby alleviate political pressure.

Both symmetrical and asymmetrical interstate wars are used in my analysis.
States facing comparatively equal or weaker opponents may have military policy options

less circumscribed than would a state facing a stronger antagonist. The balance of

 

18It is not assumed that all domestic political actors have instantaneous access to all battlefield events,
especially in democracies, but it does imply that information is diffused after some lag in time.
'9 Even relatively recent and significant conflicts such as the Sim-Vietnamese War as well as the Soviet

invasion of Afghanistan both proved very difficult in obtaining the needed information to code the requisite
data.

29

relative capabilities also engenders different sets of expectations in terms of prior
conceptions of what the outcome of a war is likely to be (Blainey 1988, Waltz 1957).
Upon fashioning baseline expectations as to the likely outcome of a war, new information

then provides updates in expectations for a wars net utility.20

Table 1: Cases and Number of Observations

 

 

 

Month- Reference .

War Observations State Opponent Conflict Type
World War II 60 Germany lFrance —+ UK Jul-40 —r US Jan-42 Autocracy-Symmetrical
World War II 44 Japan US Autocracy-Symmetrical
World War 11 48 31:3: Germany Autocracy-Symmetrical
World War II 48 Germany Soviet Union Autocracy-Symmetrical
Iran-Iraq 96 Iraq Autocracy-Symmetrical
Iran-Iraq 96 Iran Autocracy-Symmetrical
World War II 18 ermany Democracy-Symmetrical
World War II 40 ermany Democracy-Symmetrical
World War II 44 apan Democracy-Symmetrical
Korean 32 US Autocracy-Asymmetrical
Vietnam 90 S Autocracy-Asymmetrical
Vietnam 90 orth Vietnam Democracy-Asymmetrical
Korean 36 orth Korea —r China Nov-50 emocracy-Asymmetrical

 

 

 

 

The arrow symbol ‘—»’ denotes the transition from one major opponent to another during the war.

As the above graph illustrates, the wars selected for my dissertation represent
cases that are extremely violent and relatively long in duration. Acknowledging the

limitations of time and resources, the above cases should provide a glimpse into the

 

2° These wars are disaggregated into the dyadic conflicts between the principal combatants within a war,
comparable with Stam’s identification of cases (1996). Caution is taken to address the relatively minor
contribution of some states in a war (many of the UN participant states in the Korean War offered minor
assistance) and therefore weaker states in a coalition are only included if they are engaged in a
commensurate effort to that of the principal state they are aligned with. For example World War [I would
be coded as a war fought between the United Kingdom and Germany as well as another series of battles
fought by Soviet Union and Nazi forces. In the case where there is a major power fighting in a conflict
allied with other states in a coalition, the principal state of that coalition will be the subject of analysis.
Once United States joins the conflict they become the key country referenced for analysis in the study.

30

examination of the dynamics of political pressure for major wars of the latter part of the
Twentieth Century. While the following study is limited in its ability to address wars of
short duration, there is substantial amount to be learned from protracted wars that is
applicable to these more minor conflicts.

The opposing states participating in the conflict in a given calendar month are
coded as an observation. Obviously some of the cases listed above include a number of
states in a coalitions arrayed against one another. In the case of two major opposing
coalitions, the principal combatant state in one group is the referenced state. These major
combatants are then coded as an observation for every directed dyad month. For example
the two months of the F ranco-German portion of World War II there are a total of four
observations, with German observations directed at France and vice versa over both

calendar months of their mutual conflict.2|

Plan of Dissertation

The intention of the following chapters is to combine the superior components of
both qualitative and quantitative methods in order to evaluate the political pressure
theory. The case studies will offer an examination as to what factors are the most
important in determining changes in war policy. The assumption that an executive’s

constituents assess war policy previous to the outcome of a conflict is also evaluated, as

 

2' In order to account for the impact that more minor states have in terms of capabilities and casualties these
states are included throughout the range of time in which they were participants in the war.

Minor states included are Italy, Hungary, Romania and Finland for Germany in WWII. North Korea for
China in the Korean War after the entry of the People’s Republic in the war. South Korea for the United
States in the Korean War. South Vietnam for the United States in the Vietnam War. The UK for France in
World War 11, after France’s defeat UK is the major opponent of Nazi Germany on the Western Front.
Upon the US entry into World War II the UK is the minor state aligned with the US. In the Pacific theatre
of World War II the Australia and New Zealand are the minor states aligned with the US and also reflect
the UK’s contribution in the war against Japan.

31

are the casual linkages between the formation of political pressure and adjustments in
policy or deselection. The qualitative analysis will attempt to improve the validity of the
casual linkages through a process tracing analysis along with an evaluation of alternative
hypotheses.

The quantitative chapters will seek to improve the external validity of the study.
By evaluating a number of competing hypotheses and controlling for relevant variables,
the empirical significance of the political pressure hypothesis can be evaluated.
Hypotheses found to be statistically differentiable from zero are then assessed in terms of
their comparative importance in explaining changes in war policy.

Beginning with the autocratic cases, chapter three illustrates the key concepts and
examining if discontent with executive war policy lead to the emergence of political
pressure. Upon fashioning these negative opinions of a war’s relative utility, do policy
alterations and repression serve as a means to alleviate elite discontent in an autocracy?
Alternatively instances when political pressure does not emerge are also evaluated.

Chapter four qualitatively examines the democratic cases. Examples of when the
executive’s war policies become unpopular are examined in order to assess what policy
adjustments are made to accommodate political pressure. As with the autocratic cases,
incidences of when political pressure does not emerge are also assessed. This chapter
also analyzes the means by which a state is able to evaluate hypotheses regarding
alteration in negotiation positions of warring states.

Chapter five employs quantitative methods to evaluate competing hypotheses for

why changes in a state’s military strategy occur. A number of alternative hypotheses for

32

changes in negotiation position and military strategy are tested for both their statistical
salience and substantive effect.

Chapter six summarizes the findings from the previous chapters. The chapter also
triangulates the findings of the qualitative and quantitative research in order to enhance
the validity of the findings. This final chapter also highlights other potential research

topics and implications for public policy.

33

Chapter Three: Autocratic Case Studies

The following chapter examines the autocratic cases selected for evaluation and
the methodological approach designed to evaluate my dissertation’s theory. The
implication is that the casual links in the chain of logic follows the above theory and that
alternative hypotheses can also be evaluated. This chapter will also provide greater
definition to the concepts used in the quantitative chapters below.

The executives represented in the autocratic cases selected for study often faced
challenges to their position at different points in their tenure as leaders. However these
challenges are assumed to be more frequent or intense during a war, given that the stakes
are comparatively salient relative to other policy areas.22 While autocracies have a
relatively small numbers of effectively enfranchised elites, these actors may still have a
diversity of interests that drives varying cost-benefit assessments of war policy.

Assessments of a war’s utility after onset will potentially lead to demands for
changes in war policies. As domestic political actors assess the costs and benefits of
further fighting it is occasionally the case that a substantial cohort will pressure a state to
increase mobilized resources, lobby for a more aggressive strategy, or demand higher
issue-goods benefits. Generally the modal situation is the opposite. These leaders likely
ascribe more utility to an ongoing war or esteem sunk costs more highly than do their
principals.

Domestic political actors less personally invested in the conflict, having less

ideological affinity for the executive, or managing policy portfolios whose interests suffer

 

22 Evidence to date suggests that leaders do not face greater hazard of being deselected after a war than
after an international crisis (Chiozza and Goemans 2004).

34

during a war may assess the net utility of a conflict very differently. As the expected
time until benefits will be realized lengthens, the probability of achieving the issue-goods
at stake declines, and costs grow; political pressure for changes in war policy will likely
form.

The acquiescence by domestic political actors to mobilize is contingent upon a
cost-benefit assessment prior to mobilization. Beliefs as to the utility of a war are altered
by fighting as the true levels of capabilities and resolve of the opponents is evidenced.
Heterogeneity in terms of how the executive and the elites within an autocracy assess
battlefield outcomes gives rise to disparate expectations as to a war’s utility and the best
policies moving forward.

Autocratic domestic political actors are a comparatively small group. The
preferences of this group may have a lower variance than a society with broader
franchise. Empirically substantial variance as to the preferences of domestic political
actors is observed. Different elements in a states bureaucratic organization such as in
Imperial Japan can engender alternative perspectives as to policy preferences. In North
Vietnam elites positioned at different points of an ideological dimension in varying
proximity to the executive can also alter policy preferences. Domestic political actors
who are effectively enfranchised but are not a part of the executive’s coalition might also
have alternative assessments as to how best to conduct the war, as was the case with the
Iranian or German officer corps.

A major source of reduced utility to domestic political actors is the potential for
regime change either by internal revolution or through foreign imposition. Heavy costs

borne by the broader public can lead to a failure of the populace to continue to mobilize

35

for war or revolt. As domestic political actors begin to envision an existential threat to
the regime this would end their ability to collect private goods and likely increase the
prospects of punishment be it exile, imprisonment or execution. Similar to the prospect
for punishment that is thought to drive the executive to employ risk acceptant policies,
autocratic elites are strongly incentivized to examine war policies before its conclusion.

Assessments by autocratic elites that regime change is becoming more likely can
result in lobbying for an alteration in war policies or catalyze attempts to deselect the
executive. Because of an autocracies repressive capability hazarding an attempt to
supplant the executive often invites punishment, therefore typically it is only when
domestic political actors perceive a substantial loss of utility that deselection is
attempted.

These different cost-benefit schedules conditioned by ideology or role can
manifest themselves in alternative interpretations of battlefield events. The Japanese
defeats in the Marianas Islands helped spark markedly reduced assessments of the war’s
utility amongst domestic political actors who assessed the costs of further fighting more
dearly than did the executive. The crippling defeat at Stalingrad also convinced many
German officers that current war policies were inefficiently using resources and believed
Hitler was leading the state into disaster.

As the battlefield dictates the likely utility of further fighting, current war policies
are compared with plausible alternatives. If negative assessments of a war’s utility are
formed this creates political pressure to alter a state’s war policies. Autocracies do not
usually possess defined institutions and conditions by which votes are conducted with

majority or plurality decision rules.

36

The sufficient number of domestic political actors to threaten an executive’s
position is somewhat speculative contingent upon the individual case. Political pressure
is also a matter of degree, lower amounts reflective of lobbying or suggestions of
alternative policies. Greater amounts being more intensely felt and more broadly held
manifest itself in acts of dissent and violence.

The ability of an executive to repress autocratic elites is a major source of
variation between cases, permitting an executive to force domestic political actors into
submission and allowing for an alteration of the size and composition of the enfi'anchised
elite. Following the successful application of repression an executive becomes better
able to maintain his/her position and forward their preferred war policies by cowing
opponents. In order to understand how the wars are fought, the relative ability of an
executive to repress dissent must be taken into account.

Issue-goods position, the level of mobilized resources, and potentially strategy
might be altered on the basis of a rational assessment of battlefield events. Another
interpretation is that an executive, particularly in an autocracy, would be able to enforce
his/her preferred policies even in the face of active elite dissent.

Updated expectations of the likely outcome of a war can generate changes in war
policy by altering the executive’s personal preferences or collectively elites advise
changes of how the war is being conducted. In either case an essentially unitary
reevaluation of war policies would serve as a sufficient and parsimonious explanation.
However many changes in war policy, particularly major shifis, appear to coincide with
the emergence of political pressure as an executive attempts to secure his/her political

position.

37

An alternative hypothesis as to the source of strategy change is based upon the
government assessing key variables and these quantitative assessments serving as the
catalyst for change. This hypothesis developed by Gartner indicates that shipping losses,
troops, and other variables led to the adoption of alternative tactics in order to mitigate
future losses (1997). While the cases appraised were comparatively powerful
democracies, this theory’s broad contentions are also presented in the autocratic case
study chapter.

A dominant assumption of the institutional literature is that wars are examined by
domestic political interests, but only after the conclusion of a war. Anticipating the
prospect of sanction based upon the ease of deselecting a leader, the executive will only
elect to conduct wars that mitigate the likelihood of being replaced (Bueno de Mesquita
et al. 2003). Because democratic leaders are easily replaced and the public demands that
wars are fought for public security goods rather than private goods, democratic leaders
are especially selective when deciding which wars to fight (Reiter and Starn 2002).

The leaders of semi-autocratic states appear to be the ones more prone to taking
greater strategic garnbles in hopes of resurrecting their flagging military fortunes.
Because leaders of intermediate autocracies anticipate punishment at the end of a conflict
whether they suffer a minor loss or a major one, they will risk greater destruction in order
to resurrect their flagging fortunes. Changes to more aggressive war policies might then
be representative of an attempt by an executive to redeem victory at the hazard of an even
more catastrophic defeat.

The political pressure theory as to why the following autocratic states conduct

wars in the way that they do is compared with the above alternatives. The chapter

38

analyzes the selected cases on the basis of the strength of each regimes dictator. The
ordering of the cases is done with the intention of moving from the semi-autocratic states
to the more completely authoritarian. More dictatorial regimes should be more likely to
behave as if a unitary assessment of war policies is taking place or that an executive’s

constituents do not assess war policies previous to the conclusion of a conflict.

Japan: World War 11

Japan during World War II had a government that was not nearly as totalitarian as
its contemporaries in Germany and the Soviet Union; instead politics in Tokyo featured a
number of government institutions with influence. The imperial house, the military,
high-ranking members of the bureaucracy/cabinet members, former executives, and those
presiding over the repression of the broader public created a number of interests who
could attempt to supplant the executive.23

In years leading up to the war traditional political parties that had competed for
power in the 1920-305 were supplanted by the Imperial Rule Assistance Association,
characterized as “Japan’s answer to the mass political parties of the fascist states.”
(Jansen 2000, 631). A series of governments had fallen and a number of assassination
plots, usually originating in the military, undertaken as Japan expanded into Manchuria
and then the China proper (Toland 1970, 34-53). Japan’s floundering democracy

collapsed with militarist interests emerging to form a broadly composed oligarchy.

 

23 Political, military, and bureaucratic elites as being the principals actors in pre-war Japan. Political elites
consisted of the monarchy and legislature the upper house of which was composed of the gentry. The
lower house was elected but franchise limited to those paying substantial taxes to the central government
(Fukai and Fukui 1992, 27-28).

39

In October 1941 the Japanese elevated an army man, Hideki Tojo, to take the
mantle of leadership. Tojo endorsed the aggressive moves to attack European colonial
holdings and had a more hawkish predisposition in dealing with the United States.24 As
“first among equals” the Japanese Prime Minister faced difficulties coercing compliance
with war policy, despite efforts at consolidating power during the conflict (Shillony 1981,
56-64).

Imperial Japan did have effective controls over the media. Tojo was successful at
limiting access to information to the broader public or having popular dissent articulating
alternative policies (Butow 1954, 73).” By contrast Japan’s military and civilian elites
were privy to making an assessment of a war’s net utility based upon the events of the
battlefield. Further these domestic political actors could not be systematically repressed
given the checks and balances imposed on the executive’s authority.

War to secure autarky was considered a risky but acceptable option to most
domestic political actors (Russett 1967, 96-98).26 The initial cost-benefit calculation that
ratified this decision was not especially promising even amongst those who endorsed the
impending conflict, but the war received broad support amongst domestic political

ECIOI'S.27

 

2" “Tojo was the first person to be Prime Minister and Army Minister concurrently. This combination of
portfolios gave him considerable power in both military and civilian affairs. In addition, during the first
four months of his cabinet, Tojo was also Home Minister and, as such, in charge of internal security during
the crucial first months of the war.” (Shillony 1981, 10).

’5 “Deceiving the masses, the propagandists failed to confuse the more informed of the ruling elite—men
whose understanding of the actual situation was now so well founded that they could not be fooled any
longer, if, indeed they ever had been. The difficulty was that those who held official positions were
fighting not only the battle of the homeland but also the battle of their individual destinies.” (Butow 1954,
73).

26 In fact it appears to a large extent early policy decisions were not unitary and the military and other
interests spurred on the decision for war (Russett 1967, 89-90).

27 Many others believed that the war could not be effectively prosecuted but elected to support the war
despite that. The misperceptions of the United States by the Japanese leadership led to what they believed

40

Dissenters amongst the elites were represented by cabinet officials who endorsed
negotiations (Kase 1950, 38-41). Additionally the navy offered some resistance to a fight
that would bring Japan into direct confrontation with the US, although it eventually
acquiesced to war (Snyder 1992, 143). Perhaps most consistently, retired politicians and
high-ranking civilian bureaucrats believed war with America invited destruction (Kase
1950, 51, 100).

The highly salient issue-goods at stake in the conflict and the attendant risks made
for differentiated opinions as to the utility of the Pacific War at the outset. The pluralistic
composition of elites in Japan contrasts with the relatively concentrated assessment by
elites regarding the utility of a conflict at its outset. The differentiated assessments of
executive war policy began early in the conflict and grew steadily throughout. Whereas
in more totalitarian contexts it is only in the aftermath of a deepening crisis that the same
level of political pressure becomes apparent.

Despite the greater diversity of opinions regarding the war the early results of the
conflict appeared to vindicate the executive’s decision to mobilize and those domestic
political actors, predominantly military officers, who endorsed challenging the US. The
British were put to route in Malaysia and the fortress of Singapore reduced. The
Philippines had fallen and a large American force was captured, as well as the
incapacitation of a substantial portion of the US Pacific Fleet at Pearl Harbor. But these
victories did not assuage the doubts of dissenting domestic political actors.

In the central Pacific, 4-7 June 1942, the seemingly invincible naval aviation arm

of the fleet came to grief against an inferior force of Americans. The popular press

 

was still a rationale decision to engage a massively powerful opponent. In particular there was a mistaken
assumption that the US could not stomach a long conflict (Russett 1967, 96-98).

41

reported the glorious victory of the Japanese fleet, the truth of the Midway disaster was
not concealed to the domestic political actors both civilian and military. The battle had
an immediate impact upon many domestic political actors as they were disabused of
hopes for a short conflict with a supplicant United States requesting a negotiated peace.
No longer could the Imperial Navy inflict significant blows upon their capable opponent,
instead the strategic initiative passed to their enemy.

Fear of an inexorable defeat of Japan began to emerge from diplomatic and
bureaucratic elites. Believing that disaster awaited the current executive policy, a
conspiracy formed immediately following Midway, to marginalize Tojo and jumpstart
negotiations with the Allies (Butow 1954, 14). As with the other cases such as the
German invasion of the USSR or the Tet Offensive, the battlefield offered compelling
evidence that executive war policies were comparatively detrimental when compared
with other alternatives. Beginning with the portion of domestic political actors already
less favorably disposed to initial mobilization attempts to explore policy alternatives
grew.

Still, by mid-1942 Tokyo had gained far more than they had lost and it was not
the executive’s belief that the Japanese should sue for peace as current policies enjoyed
the support of many domestic political actors. Even though it was recognized that the
chances of aggregating the resources of their conquered territories into an economy that
could challenge the United States were very low (Butow 1954, 10-11).

Militarists that identified more closely with the hawkish executive understood the
implications of the Midway disaster. But even as the likelihood of defeating the

American militarily declined, high-ranking military officials began stressing Japanese

42

e’lan as a means to overcome their disadvantage in capabilities (Butow 1954, 11-12).
This cohort of hawkish domestic political actors’ continued to support more salient issue-
goods negotiation positions and greater mobilization throughout much of the war. The
fact that the executive still enjoyed broad support from the greater part of elites in
Japanese society meant that Tojo’s preferred policies continued to be broadly employed.

While supported by many military officers, fissures in Tojo’s support grew larger
as the Allies demonstrated their ability to overcome the logistical challenges of the
Pacific campaign. With the loss of the strategic initiative after Midway and sharp naval
defeats around Guadalcanal (August 1942-February 1943), the Allies were isolating
Japanese garrisons from reinforcements and supply. As the home islands were sundered
from these conquests an appreciable decline in the utility of the war was becoming
apparent to a larger group of domestic political actors.

At first reticent to join in the lobbying for war policy change, Emperor Hirohito

increased his involvement and began to organize critiques of Tojo’s policies.

According to Kido’s [Lord Keeper of the Privy Seal and counselor to Hirohito] diary, after the
defeat of Guadalcanal in 1943 Hirohito remarked: “Strangely, the war is lasting too long.” Later
he supported Foreign minister Shigemitsu’s idea that a special envoy be dispatched to Moscow to
negotiate a peace, but he did not insist on it when the army objected. When the fall of Saipan
became imminent, the emperor made an unusual move by summoning the Board of Field Marshals
and fleet Admirals, a prestigious body with little power that had rarely been convened. . .The
Board recommended the setting-up of a new line of defense in the Pacific and the unification of
the air forces of the army and navy. Shillony 1981, 60.

The military results demonstrated a reduced likelihood of obtaining a favorable
settlement form the Allies. Hirohito lobbied to initiate negotiations with the United
States via the Soviets in order to realize benefits more quickly and secure more limited
issue-goods while Japan still could. Further the emperor also suggested an alteration in
military strategy might provide for a more robust defense of the home islands. The fears

of Japan’s decreasing ability to resist were soon realized.

43

Throughout 1943 the protective belts of imperial conquests surrounding the
Japanese home islands began to be penetrated. The sense that executive war policies
were exaggerating an already difficult situation increased political pressure amongst new
cohorts of domestic political actors. Tojo’s was apparently unable to mediate inter-
service rivalries and effectively allocate mobilized resources. Detractors that had
principally been composed of foreign-service and civilian elites now included members
of the monarchy and military (Shillony 1981, 58-59).28 Growing political pressure
coalesced as factions interested in a negotiated settlement combined their influence.

The decisive setbacks against the Allies meant that Japan’s domestic political
actors took an increasingly negative view of the war’s utility generally and the
executive’s conduct of the conflict. Without a meaningful capacity to repress dissent,
political pressure became a major consideration in determining how Japan would
prosecute the war. Ignoring dissent invited the prospect of deselection and
accommodating political pressure meant detraction from an executive’s favored policies.

Regardless the substantial impact of differentiated assessments of a war’s utility
would constitute a meaningful shift away from a unitary assessment of how to conduct a
war. Either by altering the executive whose influence substantially guides policy or by
altering war policy to mitigate political pressure results in substantively different

outcomes.

 

28 “Major-General Prince Kaya, cousin of the empress. . . had reservations about Tojo. On 10 December

1942 he told his uncle that the general staff of the army blamed Tojo for failing to provide the ships that
were needed for the defense of Guadalcanal. In July 1944 Prince Kaya became alarmed about the war
situation and told his uncle [Hirohito] that unless Japan sought peace, she would face disaster. He
suggested that the three imperial princes who held senior ranks in the army. . .appeal together to the
emperor to dismiss Tojo and appoint Prince Fushirni as the over-all commander of the imperial forces.”
(Shillony 1981, 58-59).

44

Strategy became an increasing source of reduced utility for domestic political
actors.29 Though most high-ranking officers appeared to garner more utility from
fighting than did their civilian counterparts, strategic missteps were causing these
domestic political actors to take issue with executive policy. Japan’s merchant and
military shipping losses to American submarines were extraordinarily heavy, resulting in
difficulty resupplying garrisons and keeping arms industries functioning as raw materials
were chocked off (Butow 1954, 22.23).30

Admiral Takagi’s report, made in early 1944, pointed to merchant and military
shipping situation as being especially precarious. In response to the situation he extolled
the executive to seek immediate efforts at a negotiated peace or witness Japan’s complete
destruction (Toland 1970, 476). Tagaki contemplated what might be suitable negotiation
offers for the Allies, taking council with other elites it was decided that withdraw from
mainland China and Manchmia would be a basic starting point. The issue-goods that
these domestic political actors still hoped to extract were political or at least economic
concessions in Taiwan and Korea in order to provide for the raw materials needed in the
Japanese proper (Butow 1954, 21).3 I

In spite of his military credentials, controlling the powerful military bureaucracies
was difficult for the prime minister (Russett 1967, 89). Tojo’s inability to deal with
shipping losses to submarines constituted an existential threat to the Japanese war effort.

As the costs of the war remained high and the benefits of further fighting declined

 

29 Sensing this growing political pressure Tojo reconstituted his cabinet in April 1943 (Kase 1950, 68).

3° The response of the military, or there lack of, contrasts sharply with the one of Britain’s similar situation
in the World Wars in which the military embraced new tactics to deal with the losses at sea (Gartner 1997).
The effect of shipping losses to Japan is difficult to understate and the lack of progress tactically was
equally difl’icult to comprehend.

3' Takagi’s grievances with current war policies became so fundamental that he later conspired with other
naval officers to assassinate Tojo in order to remove the obstruction to peace talks (Butow 1954, 21).

45

precipitously, the diversity of domestic political actors dissatisfied with executive
performance increased. Throughout 1944 a deteriorating military situation fueled
lobbying for policy alternatives and a growing conspiracy to deselect the prime minister
(Toland 1970, 508-509).32

The decision to mobilize for war was aided by the assumption that the Americans
were not resolved enough to fight a protracted struggle (Russett 1967, 99). This
assumption helped fashion the attritional strategy executed after the initial weeks of the
war, designed to wear down American resolve through a drawn out conflict over great
distances. By 1944 fighting a conventional war against the Americans had revealed
Japan’s material inferiority and produced increasingly lopsided defeats. No more so than
in the defense of Saipan.

The Marianas Islands were close enough to metropolitan Japan to serve as a base
for strategic bombing as well as a major step forward toward invading the home islands.
In an effort to challenge the US fleet supporting the invasion of Saipan, a strike force
composing the Imperial Navy’s residual carrier force attacked. The June 1944 battle
would become known as the Marianas Turkey shoot.

Japanese naval aviation equipment and training was now seriously deficient
resulting in a crushing loss. The resources being devoted to the war were facing

diminishing marginal returns in terms of the likely issue-goods settlement that could be

 

’2 “. . .Prince Konoye [naval officer and member of the monarchy] had become the confederate of a score

of military as well as civilian leaders—among them General Koji Sakai of the General Staff and Admiral
Okada—who were disturbed by the course of the war and Tojo’s leadership. General Sakai made a
clandestine visit to Konoye’s suburban home. ‘To be on the safe side’ the general wore civilian clothes. .
.’Germany still has defensive power, and while the enemy has to fight in both east and west we should take
advantage of the situation and enter into negotiations for peace. It will not be to our advantage to wait until
Germany is defeated.’ Tojo could not possibly negotiate such a peace. A new cabinet must be formed.”
(Toland 1970, 508-509).

46

expected. Directly challenging the United States and grinding down American resolve
over the vastness of the Pacific had failed (Toshikazu 1950, 78; Frank 1999, 89).

This illustrates one of the more frequent aspects of domestic political actors and
the executive assessing a war’s utility differently. Tojo as partner in the move to secure
economic resources through war and architect of military policy since had a stronger
interest in continuing to prosecute the conflict than did a growing number of his
principals. Not only did the executive have a more hawkish orientation to begin with, but
the personal prestige expended upon the war meant that he obtained greater utility
through further fighting than did his constituents.

The contention between elites and domestic political actors was never undertaken
with regard to the necessity of total mobilization and the attendant costs of a total war.
Instead the relative efficiency by which mobilized resources were being employed and
the expected issue-goods benefits that could be attained was the source of political
pressure. Most important the sense of urgency amongst Tojo’s principals to end the war
as quickly as possible in order while Japan retained the ability to exact costs from their
opponent.

Compelling an executive to accept the broaching of negotiations or a modification
in the issue-goods position that a state takes is one of the most frequent implications of
political pressure. However, in the case of World War II Japan an accommodation of this
pressure was not reached. The continued intransigence of the executive’s to pursue
negotiations engendered increasing political pressure.

Both civilian and military dissidents now saw that executive policy was deeply

antithetical to their interests and transitioned from lobbying for policy change toward

47

direct efforts to deselect the executive.33 While Tojo used his political acumen to defuse
previous attempts to supplant him, the military developments laid bare the decreasing
ability of Japan to garner concessions from the Allies. As the plight of the regime
became increasingly clear, the level of political pressure also increased. A greater
portion of Tokyo’s elite began to view current war policies as deeply antithetical to their
interests.

Fears as to the regime’s future was deepened by as the relative cost of Japan’s
resistance to the Allies was soon to explode. The fall of Saipan foretold massive
bombing of the Japanese proper, fiirther escalating the likelihood that the public’s will to
mobilize might snap. A critical mass of dissenters formed leading to the executive’s
resignation.34 The declining probability of attaining anything close to the objectives at
the war’s onset was not addressed by Tojo. The failure to adapt military strategy did
nothing to improve most domestic political actors’ utility calculation and without the

ability to repress elites Tojo was deselected.

 

’3 Saipan was widely recognized as the point at which most domestic political actors began to appreciate
the how seriously the reduction in issue-goods position would have to be in order to achieve a settlement.
Leading to heightened domestic political pressure culminating in the deselection of Tojo. (Yangana 1949,
616-617).

3’ “The cabinet had nearly failed in June [1944] when two cabinet ministers had threatened to resign. Tojo
had managed to counteract this move by reconstructing his cabinet, but the Saipan invasion, which came
immediately afterward, was a crisis of much greater magnitude. The loss of this island lying far within
Japan’s defense perimeter, the disintegration of Japan’s front in Burma, and the magnification and
multiplication of problems on the home front proved to be too much for the cabinet to handle. While
several ministers pooled their efforts to sabotage the cabinet from within, the senior statesmen worked to
destroy it from without. Finally, on July 17, 1944, the intrigue came to a head. Quite informally and on
their own initiative, the former Premiers met at Baron Hiranuma’s residence to clarify their stand and to
forestall an attempt by Tojo to maintain himself in office by resorting to another cabinet shuffle and
reorganization.” (Brooks 1968, 27-28).

48

Figure 2: Distribution of Elite Opinion in Japan

Beginning of War After fall of Saipan
Executive’s policy . . . . , . Distribution of
preferences/ Distribution of Executive 5 policy Domestic Political
current war policies Domestic Political preferences/ Actors’ preferences

\ Actors’ preferences current war policim

 

 

 

 

 

 

4 > e >
Less favorable More favorable Less favorable More favorable
relative utlllty relative Utility relative utility relative utility

Upon the removal of Tojo in July 1944 Kuniaki Koiso, another general, was

elevated to the position of prime minister. Koiso’s selection was a consensus choice

made in part because he was considered somewhat more amenable to a negotiated peace

than his hawkish military peers (Toland 1970, 603). The new executive was not

burdened by being the initiator of the conflict and was expected to be less disposed to

taking a hard-line in terms of negotiations. Instead, as was his predecessor, the new

prime minister appeared to garner more utility from a distended conflict in the hope of

obtaining a better issue-goods settlement than did most of Japan’s domestic political

actors.

Despite being isolated from many military decisions, Prime Minister Koiso put

his own credibility into a defense of Leyte in the Philippines. A diversion of a large

49

portion of Japan’s mobile army formations and serviceable navy were sent to the
archipelago for its defense. Tokyo’s naval and army units on Luzon Island were
subsequently crushed by a vastly stronger force. To many domestic political actors it was
a proliferate expenditure of precious military resources, especially given the dismal
results for what the Koiso billed as an apocalyptic struggle (Frank 1999, 90). The defeats
in and around the Philippines continued to point to a limited ability of Japan to resist the
Allied demand for unconditional surrender.35

As the attritional strategy employed against the Americans was becoming even
more futile. Military commanders, with the support of the new prime minister, began
pursuing a punishment strategy. In order to exact a heavy toll from American forces a
dedicated suicide formation was created known as the “Special Attack Corps” (Toland
1970, 631)}:6

Beginning in January of 1945, at the Battle of Leyte Gulf, and with much greater
numbers and effect in the defense of Okinawa, the Japanese began using kamikazes to
inflict heavy losses upon the US Navy. Whereas the use of planes and pilots in
traditional engagements at Luzon and the Marianas resulted in nearly complete losses
with little appreciable gain, the suicide attacks inflicted heavy damage on American
shipping. An alteration in strategy used the resources mobilized to much greater effect
and consequently appeared to enhance the bargaining position of Tokyo versus

Washington.

 

3’ Originally stated in Cairo 1943, this was followed by other ultimatums and finally the Potsdam
declaration of July 1945.

36 The decision to use suicide aircraft, including the emperor’s naval attache rather than the unsuccessful
conventional attacks, the efficacy of these attacks was shown to be very impressive (Hastings 2007, 165).

50

The alteration in strategy did increase the net utility of continued fighting
(Slantchev 2003b)”, however Koiso’s incompetence in managing the bureaucratic
interests of the military was patently clear. The executive’s continuing reticence to
seriously pursue negotiations and the crushing defeats of some of Japan’s best units, led
to Koiso’s removal in April 1945.

Koiso’s replacement the retired admiral Kantaro Suzuki was cagey with respect to
how diligently he would pursue a peace. This ambiguity was later removed when he was
quoted as saying that “one of the hundred million people to cast away what holds them
down to become glorious shields for the defense of the nation” (Toland, 1970, 864). This
position appealed to the minority of elites mainly constituted by the hard-line military
officers who sought a cataclysmic struggle on the home islands rather than surrender.

But for the majority of domestic political actors more favorably oriented toward
negotiations this was not at all comforting.

The likelihood of having to surrender unconditionally to the Allies was now
trumped by the possibility of an internal revolution. Beginning in February 1945 massive
bombing raids increased the costs of the war greatly in terms of lives lost, destruction of
urban centers, and a sharply decreasing standard of living. The monarchy and other
conservative domestic political actors read the signs of a possible leftist revolt (Brooks
1968, 17-18).38 Ironically it was the Soviets who were viewed as a potential candidate

for mediation with the United States (Butow 1954, 84).

 

’7 Japan despite knowing the ultimate outcome of the war continues to fight in order to inflict costs upon
the opponent in hopes of obtaining a better issue-goods settlement (Slantchev 2003b).

3’ “Early in 1943, for instance, Kido [Lord Keeper of the Privy Seal and counselor to the Emperor] had a
three-hour discussion with Prince Konoye [a prominent member of a royal family], who was extremely
pessimistic about the war situation. During the course for this long conversation. . .Konoye repeatedly
spoke of the necessity of terminating the conflict as soon as possible lest unsettled internal conditions lead

51

Even as early as 1943 it was feared that the costs of continuing the war might lead
to a communist takeover, threatening the positions of privileged for domestic political
actors most especially the monarchy (Butow 1954, 48). The long simmering discontent
amongst the public was soon to be brought to a boil in 1945. The domestic political
actors, with the moral currency of the throne, began to desperately seek negotiations with
the Soviets, as Moscow was still officially neutral toward Japan.39

The Foreign Minister Togo thought there was a possibility of a mediated
settlement via the Soviets, but realized that Moscow would demand a dear price for its
aid.40 The negotiations with the Soviet Union did not get off to an auspicious start, as the
defeat of Germany in May 1945 had left them in a stronger position to expand their
influence in Asia. More importantly, Prime Minister Suzuki still did not seriously
entertain a negotiation position that was realistically going to be accepted by Moscow.

The terms of the Potsdam Agreement of July 1945, reiterated the unconditional
surrender of Japan as the Allies only acceptable terms. Foreign affairs bureaucrats and
other civilians were joined by a large portion of high-ranking military officers to

provisionally accept the terms of surrender. But the executive backed by more hard-line

 

to an intensification of Communist activity within Japan. The issue, in Konoye’s mind, was clear-cut; end
the war now or be prepared to see communism emerge as the ultimate victor.” (Brooks 1968, 17-18).

39 Some of the more hawkish domestic political actors, actually believe that there was an opportunity to
form a military alliance with the USSR (Toland 1970, 754). Although more experienced officials
accustomed to dealing with Moscow were much more realistic (Frank 1999, 221-222).

’0 “Foreign Minister Togo warned that in order to be successful in this venture Japan would have to be
prepared to make substantial concessions. He even went so far as to suggest that the Empire might have to
return to its pre-Russo-Japanese War boundaries——a retrogression which would have amounted to a
cancellation of the Treaty of Portsmouth and of the economic concessions which were subsequently
negotiated on the basis of that treaty. Since it was probable that Moscow would lodge a claim to North
Manchinia, Togo openly spoke of relinquishing that area to the Soviet giant.” (Butow 1954, 84).

52

domestic political actors attempted to carve out a few concessions in a peace deal
(Toshikazu 1950, 231).41

The battlefield again illustrated to domestic political actors the futility of further
resistance. The Soviet invasion of Manchuria not only showed how weak the Japanese
army was, it also eliminated the only plausible candidate for reaching a mediated
settlement short of American terms. Equally shocking was the atomic explosions while
less damaging than other bombing attacks, revealed the devastation Japan invited by not
seeking immediate resolution of the war.

Even in the midst of these catastrophic events, infighting in this last of the
Japanese war cabinets pitted those advocating a modest modification of unconditional
surrender against the hardliners (Toland 1970, 925). It was only when the emperor
himself intervened that the military leaders supporting the executive were willing to
accept accommodation with the Allies (Kase 1950, 23 5). Obviously the executive’s
decision to surrender was not a function of attempting to preserve his position in the
regime, but it was also not a unitary decision made as a reflection of the Suzuki’s
personal preferences in terms of war policy.

As the political pressure hypothesis predicts, the battlefield created different
impressions as to the utility of the war between domestic political actors and the
executive. While the choice to mobilize is assessed as a rational calculation at the war’s

outset (Russett 1967), assessments of the war’s utility began to differ greatly between

 

" “All agreed on one point, namely, the preservation of the imperial house. Should the Allies refuse that,
we had no choice but to fight on to the bitter end. But War Minister Anamic, and Umezu and Toyoda, the
two chiefs of staff, insisted on adding three other conditions. They were: 1. No occupation of the
homeland by the Allies. 2. Voluntary withdrawal of Japanese expeditionary forces and disarmament and
demobilization of our armed forces by Japan herself. 3. Voluntary punishment of war criminals, by the
Japanese government.“ (Toshikazu I950, 231).

53

agent and principals as battlefield revealed the relative capabilities and resolve of the
combatants. A greater proportion of domestic political actors employed political pressure
in order to alter war policies. As expected greater levels of political pressure when not
accommodated by changes in war policy led to the deselection of the executive.

The manner in which Japan conducted World War 11 did not reflect an objective
appraisal of battlefield events and costs and benefits. Instead of reflecting an effectively
unitary appraisal of key variables such as casualties or wins, the causal impetus behind
some major changes in negotiation position and strategy was the result of political
pressure. The failure of the executive to respond to demands for changes in war policy
resulted in the deselection of two prime ministers and attendant changes in war policies
following their removal.

A unitary assumption regarding how Japan fought the war is not as satisfying an
explanation. The cartelized politics of Japan resulted in overexpansion and invited a
balancing coalition addresses the origins of the war, but does not explain why dissent
with war policies manifests itself. If domestic political observers review the selection of
a war only after its conclusion then why are two different executives removed from
office.

War selection also does not explain why the broad mandate that the conflict
received granted by domestic political actors, but was later rescinded. The case for
mobilization before the attack on Pearl Harbor did not appear to be reflective of an
attempt to systematically deceive Tokyo’s elites, instead their appeared to be broad

consensus as to the risks and rewards for mobilization (Russett 1967).

54

The military strategy employed was surprisingly stable. Given massive
asymmetries in material capabilities, the executive continued to pursue an attritional
strategy. Moreover in the face of catastrophic losses in Japanese shipping, the executive
did not actively remedy the situation with a rearrangement of military resources or
strategy.42 The static nature of strategy appears to reflect the preferences of the
executive, at least during Tojo’s time as prime minister, even as the level of political
pressure grew.

The assumption of a more risk acceptant strategy in a war in order to avoid
punishment is only partially vindicated in the Japanese case. Tojo is deselected from
office more than a year before Tokyo’s capitulation, suggesting that his constituents were
observing the progress of the war well before its ultimate conclusion. The embracing of
suicide strategies such as kamikazes occurred at the latter stages of the war, but these
tactics originated in the military leadership who lobbied for their application by the
executive.

In terms of mobilization, Japan was committed to a total war throughout the
duration of the conflict. Both of these would suggest that while fear of punishment by
Tokyo’s executives was real enough, it was as much the failure to adapt to changing
circumstances rather than a radical shift toward a more aggressive strategy appears
evident. For the most part the executive was able to pursue their favored war policies
with respect to military strategy and issue-goods negotiation position. This is especially
striking given that the prime minister in Imperial Japan only had a very limited capability

to repress domestic political actors.

 

’2 This shows a marked contrast to Gartner’s findings regarding England’s evolution in strategy during the
struggle for the Atlantic in World Wars I and II (1997).

55

While the executive was able to dominate war policies, the lack of
accommodation to political pressure was met with deselection well before the war
concluded. This demonstrates that deselection was driven by political pressure of an
executive’s domestic political actors and that removal of two prime ministers effectively
altered what policies would be employed. Additionally, when major shifts in war policy
did occur it was when the executive was under severe duress by his principals rather than
by an assessment of objective indicators such as casualties.

The above case appears to be indicative of the political pressure model. Even
accepting the comparative simplicity of alternatives, the unitary actor assumption does
not grasp the reasons for war policy changes. Consequently, the manner in which the war
against the United States was fought by Japan and the terms under which it was settled

are reflective of more than a unitary assessment of battlefield events.

People’s Republic of China: Korean War

The Korean War began as an attempt at reunification by Kim Song-ll. The result
was a quick collapse of the South Korean defenses and what appeared to be an
unambiguous success for Moscow’s client. The unexpected decision to intervene by
Truman preserved the port of Pusan in the southeast comer of the peninsula and through
a desperate struggle kept the Republic of Korea (ROK) alive.

Receiving a Security Council mandate to halt North Korean aggression, the
American General Douglas McArthur took command. MacArthur’s aggressive
amphibious landing at Inchon north of the communist encirclement separated the North

Korean army from their logistical support. A breakout assault from the Pusan perimeter

56

followed the successful landing and Democratic Republic of Korea (DRK) faced
collapse. In October 1950 the American led UN force regained Seoul and the 38th
parallel, General MacArthur then sought to reunify the country as an ally of the US.

Repeated warnings that the Chinese would intervene were misunderstood or
ignored (Christensen 1996, 149-152). Even evidence of captured People’s Volunteers
Army (PVA) regulars amongst North Korean prisoners was not appreciated as a
harbinger of direct military intervention." While the American led UN forces pressed
forward in November 1950, the debate over intervention in Beijing had been concluded
but not without trepidation (Scobell 2003).44

The founding of the People’s Republic was announced on 1 October 1949, and
this new regime was still in the process of consolidating their hold on continental China.
Although Mao was unenthusiastic about a conflict with the United States (Gustov and
Hwang 1980, 55), the executive endorsed the war and sought to mobilize for war. The
regime’s domestic political actors were the senior military and civilian leaders of the
Chinese Communist Party (CCP). Amongst his constituents Mao’s influence over
military affairs was great, but could not be construed as hegemonic. In fact major doubts

regarding an intervention in Korea were freely expressed.45

 

’3 People’s Volunteer Army was the euphemism given to Chinese regulars serving in Korea.

’4 Mao was the pontiff of political-military thought, but other major actors in the CCP hierarchy also were
influential. Despite the security threat that loomed, many of Beij ing’s military-political elites were very
ambivalent about the prospect of a war in Korea. Others were even more adamant in their opposition
including Lin Biao an esteemed Red Army general from the Civil War. (Scobe112003, 79-94).

’5 “The cleavage between those who wanted to get on with the economic reconstruction of China and those
who saw the need to defend China’s security at the gate, was probably the deepest. While obviously not an
open fight, let alone one fought to the finish, this debate seemed to surface before the decision to intervene,
as well as during the course of the war of attrition. As with all these policy debates it is hard to identify
specific factions, let alone factional leaders, but some sort of debate did take place. The dark shroud draped
over these debates was no doubt the result of basic unity derived from recently completed civil war
struggle, and the concern with the looming US threat in Asia.” (Segal 1985, 105).

57

Not only was there apprehension with regards to the costs of the war amongst
civilian and military elites, some domestic political actors dismissed the importance of
the security benefits of fighting. While the case for mobilization was predicated upon an
existential threat to the People’s Republic if the US was allowed to use Korea as a base,
other disagreed. Some Red Army generals believed an actual invasion of China had not
yet been shown to be the objective of the American intervention. Even if the US
launched an invasion it would not necessarily have meant the end of the regime (Gustov
and Hwang 1980, 55).

CCP officials administrating economic portfolios were especially opposed to the
invasion of Korea. Chen Yun minister of finance thought the challenge of repelling
American aggression to be “incompatible” with economic reconstruction program
planned by Beijing (Gurtov and Hwang I980, 55).“ The degradation of the Red Army in
combat with the United States could facilitate a Nationalist comeback or reduce the
CCP’s ability to repress dissent (Uhalley 1988, 99). The costs that domestic political
actors ascribed to the intervention into Korea reflected fear of jeopardizing the
communist revolution.

Offsetting these debits Mao annunciated two principal objectives of the military
campaign. “First, Korea’s physical status quo must be restored, meaning that there

should be no US or ROK troops north of the 38th parallel. . .Second, Korea’s political

 

’6 Older sources depict the Korean War as being widely held in Beijing as an existential struggle
necessitating the greatest effort (W int 1960, 62).

Marshall’s Chinese contact disclosed information about the formation of Chinese leadership. Zhu points to
US sources that highlighted a slightly different factionalization of CCP elite. “The Chinese government,
according to this Chinese connection, was made up of three different factions: ‘the Moscow-oriented
Communists (Stalinists); China oriented Communists; and non-Communists.’ ‘A small inner clique,’
which in fact decided China’s intervention in the Korean conflict, kept ‘apparent, stronger’ ties with
Moscow. Although Mao was thought to be “much in the Russian camp,” he was also one of the China-
oriented Communists, acting on the basis of his perceptions of China’s national interests.” (Zhu 2001, 51).

58

status quo prior to the outbreak of the war must also be restored.” (Xia 2006, 48).
Despite somewhat less expected utility amongst most domestic political actors relative to
the executive, mobilization was endorsed. Like the Japanese case Mao did not possess the
opportunity or willingness to repress those CCP elites that had destroyed the Nationalist
regime.

With North Korean resistance crumbling, McArthur’s UN forces had become
stretched in their rapid movement toward the Chinese border. The lack of caution and
with substantial gaps between units opened up an opportunity for the PVA. A force of
260,000 Chinese troops had amassed along the border (Korean Military Institute 2001b,
109-115). The initial impact of the assault was both shocking and impressive (Zhang
2003, 98-114). The commander of Chinese forces, and effectively all communist forces
in theatre, General Peng Dehuai employed a strategy perfected in the latter part of the
recent Civil War. The tactics that had been so potent against the Nationalist forces
appeared to be just as successful against their surprised foes.

While the bulk of their forces hit the UN troops head on, Chinese forces also
infiltrated around the American and Korean units attacking their flanks and encircling
their opponents with infantry. Upon slipping past the opponent’s front lines, PVA troops
then began attacking the supply lines and severing smaller units away from their parent
formation.47 Shocked by the tens of thousands of Chinese troops and their seeming
ability to appear anywhere on the battlefield, the strung out UN forces were compelled to
make a hasty retreat. This campaign near the China-North Korean border began the first

of five offensives initiated by the Communists.

 

’7 In October 1950 Peng was also ordered to inflict maximum casualties in order to compel a negotiated
settlement from the US. (Chen 1989, 191).

59

By the end of the third winter-sprint offensive the PVA laid claim to the South’s
capitol. Success convinced the executive that greater issue-goods demands could be
garnered through fighting. 4' The expectations of a Korean Peninsula reunited under
Pyongyang and a supplicant US begging for the return of its captured soldiers were
potential benefits of great utility to a regime struggling for international legitimacy.49

The results of combat with the UN forces had left the impression upon Mao that
much more could be achieved in terms of issue-goods benefits. While the threat to the
regime seemed much decreased in the spring of 1951, this also mitigated the principal
rationale for mobilizing for the war and helped create disparate assessments of the
conflict’s utility.

Mao insisted on pressing his advantage despite the warnings of the less optimistic
theatre commander General Peng (Li et al. 2001 , 20). The un-standardized Chinese
infantry equipped with a cacophony of different firearms, scarce and assorted artillery as
well as very limited motorized transport quickly overwhelmed their logistical
capabilities. To make matters worse the extended supply lines were decimated and
exposed communist troops were mercilessly pounded by American airpower.

While road bond and easily susceptible to logistical disruption, the UN forces
were very mobile and outran their pursuers. Surrendering Seoul in January 1951 for
more defensible positions, the UN forces were resupplied and reorganized. When the
communist forces launched their last spring offensive they faced an opponent with

prepared defenses and massive firepower. Without the benefit of surprise and opposed

 

’8 Mao protested Peng’s decision not to pursue the UN forces after the third offensive, either due to a lack
of capacity or willingness Peng did not comply. (Li et al. 2001, 20).

’9 Mao would later make statements of his issue-goods position that included the Chinese seat in the United
Nations amongst other diplomatic prizes. (Foote 1990, 35).

60

by a continuous line of defense the Chinese non-mechanized maneuver strategy ground to
a halt (Harlan 1982, 47). A massive counterattack was then launched against the PVA
and the communists were steadily driven north relinquishing Seoul.

Mao’s insistence on launching the last spring offensive in early 1951 had
markedly altered the cost-benefit calculations of many CCP elites. The attacks led to the
decimation of some of the Red Army’s best formations. Moreover the enormous expense
to the already fragile economy, and the paucity of Soviet aid made the conflict in Korea
an expensive one. The battlefield now indicated that pushing the UN forces off the
peninsula was very unlikely, a change in issue-goods negotiation position reflected these
realities. The new objectives of the war were a reversion to the earlier objectives stated
by Mao, to sustain a pro-Beijing regime in North Korea (Chen 2001, 61).50

The Chinese also elected to alter tactics to a more cautious strategy, rather than
the sweeping maneuver tactics employed in the five winter-spring offensives. Had this
decision reflected the unitary assessment by Mao or if domestic political actors were not

exerting political pressure on war policies, then this would support an alternative

 

5° “The Communist defeat in the Fifth [winter-spring] Campaign forced Mao and the other Chinese leaders
to reconsider their aims on the Korean battlefield. Realizing that a huge gap existed between the capacity
of Chinese troops in Korea and the ambitious aims that Beijing had assigned to them, Mao became willing
to conclude the war short of a total Chinese/North Korean victory. In late May 1951, Beijing’s military
planners, following Mao’s instructions, conducted an overall review of China’s strategies in Korea. Nie
Rongzhen, China’s acting chief of staff, summarized the consequences of this review process in his
memoirs: ‘After the Fifth Campaign, the Central Committee met to consider what steps should take next.
The opinion of the majority is that our forces should stop at the 38’h parallel, continue fighting during the
armistice talks, and strive to settle the war through negotiations. . .Mao and his fellow Beijing leaders
concluded that the success of the Chinese troops in pushing the US/UN forces back from the Yalu River to
areas close to the 38'” parallel had sufficiently put them in a position to claim that China had already
achieved a great victory.’ Under these circumstances, the CCP leadership decided at the end of May that
China would adopt a new strategy, one with a keynote of ‘fighting while negotiating,’ and China’s.
operational aims would now be redefined as pursuing an armistice by restoring the prewar status of Korea.”
(Chen 2001, 97).

61

hypothesis. However the narrative of domestic political actors and executive in one
accord with respect to war policies seems incongruent with most evidence."

The domestic political actors were largely veteran officers in the Chinese Civil
War (Segal 1985, 105-109). This acumen gave rise to criticisms of strategy as the costs
of the war increased.52 The executive’s preferences for the maneuver style warfare
employed in the revolution, was becoming a comparatively inefficient use of resources.
The destruction of some of the most potent units in the Red Army meant they were
unavailable for suppression of residual Nationalist forces on mainland China or to defend
against a future assault by J iang Jie-Shi’s armies (Zhang 1995, 154; Uhalley 1988, 99).
Threats to the regime’s stability led a growing number of domestic political actors to
ascribe a lower utility to war policies than did the executive.

The principal reason for Mao’s reticence to alter military strategy was that the
legend of “people’s war” lay at the heart of his mandate for power. When the sweeping
offensives carried out by lightly-armed infantry lost efficacy (Jencks 1982, 47), a
retrenchment away from this romanticized warfare was not in the executive’s interests.
Debates over military strategy paralleled the ideological rift amongst the executive and

domestic political actors (Dreyer 1985, 39-40).53

 

5 l The realities of reduced expectations eventually were accepted by Mao and tactics were switched to
trying to encircle smaller UN formations rather than the grander designs of cutting off large units. (Mott
and Kim 2006, 128).
”The following passage illustrates the debate over military strategy and indicates that Peng was the
proponent of continuing the assault. “There can be little doubt that some sort of debate on military strategy
had taken place. Cultural Revolution sources suggest that the first three battles of the Korean war [until
April 1951] were fought according to Mao’s ‘operational plan’ urging active attack instead of ‘taking a
rest’. . .Peng’s reported desire for a further offensive was blocked by Mao in favor of ‘protracted fighting
and positive defense’. Obviously much of this assessment must be suspect, coming as it did at the height of
the anti-Peng campaign in 1967. Nevertheless, it does suggest that military set-backs did encourage
pragmatism and strategic debate.” (Segal 1985, 108).

3 “ While professionalists and ideologues have shown few differences on the matter of the military’s role
in politics, other issues have differentiated the two categories. Ideologues have typically given more
weight to political, as opposed to technological, factors in military training, have stressed the primacy of

62

Amongst elites in the CCP the principal ideological dimension was a continuum
between doctrinaire communists more sympathetic to Mao and professionalists who
wanted greater modernization.“ The immediate application to the Korean War was the
emphasis that would be placed upon professionalization and equipment in the military
(Kamow 1972, 71).55 Mao’s mandate for power rested upon the foundation of his fusion
of sinofied communism and military strategy that enjoyed such success in defeating the
Nationalists. Questioning Mao’s maneuver strategy reduced the executive’s influence in
that it was his military tactics and ideological teachings that lent him most of his
legitimacy.

After the devastating defeats of the last Chinese Winter-Spring offensives,
criticisms of Mao’s strategy reflected concern over the degradation of the Red Army and

the need for modernization (Powell 1968, 25 7).5 6 It is doubtful that elite critics of Mao

 

people over weapons, and are likely to place more value on the lessons of the People’s War than
professionalists. It would, however, be erroneous to assume, as some have done, that there is a very clear
line dividing one group definitively and for all time from the other. In fact, the weight of evidence
indicates that there is a continuous spectrum of opinion on how the military should be organized and
trained, and that individuals, and presumably the groups they belong to, can and do change their positions
on the spectrum according to changing circumstances.“ (Dreyer 1985, 39-40).

5" Throughout the economy and the early years of Communist rule, there was an ongoing debate between
the relative virtues of being “red” or “expert”. Meaning would deference to revolutionary leaders or party
members take control of lower-level policy making and technical tasks or were professional
bureau/technocrats to fill these positions. The latent dimension of political opinion within the CCP was
between orthodox Leninism “with the requisite Maoism and Sinofication” as opposed to pragmatism and
doctrinal flexibility. The doctrinaire believed that adherence to communist prescriptions and emphasizing a
revolutionary spirit that virtually all obstacles could be overcome. These leaders called upon the
dialectically inevitably of communism to produce optimal policy outcomes, as the emancipated proletariat
was ascribed the ability to overcome material constraints. (MacFarquhar 1997).

5’ “It would be a crude oversimplification to say that Mao totally disparaged the value of military expertise
and that Chinese Army professionals rejected his focus on ideology. . .Therefore, many of the differences,
exacerbated by related doctrinal, economic, and international problems, gradually polarized to the point at
which Mao and a significant faction within the Chinese military establishment found themselves in bitter,
irrevocable disagreement.” (Kamow 1972, 71).

56 “Since the Korean War this criticism, which has waxed and waned, has come especially from some of the
professional officers of the PLA, who themselves are party members. . .the major critics are said to have
been Chairman Liu Shao-chi, Marshal Peng Te-huai and General Lo Jui-ching, the former Chief of Staff, as
well as the influential regional party leaders and military commissars, Tao Chu and Li Ching-chuan, all of
whom have been purged. Most charges, true or false, have been made against them, but the attacks are
meant to include their supporters. . .There is some validity in the statement that the ‘fundamental’ question

63

would have been able to or even desired to deselect him. The lobbying for an alteration
in strategy was accommodated by the executive in part out of fear that he might suffer
even greater degradation in political authority than would a repudiation of people’s war
(Zhang 1995, 154; Segal 1985, 110).57 It was a recalcitrant executive who was
compelled by his domestic political actors to alter strategy, rather than pursue his own
unitary preferences (Mott and Kim 2006, 128; Li et al. 2001, 20).

Divergent assessments of the relative utility of executive war policies meant that
critiques of executive policy were the subject of lobbying efforts by some CCP elites.
What changed the dynamic in their favor was the increasing apprehension of many
domestic political actors as to the underlying position of the young republic. The number
of domestic political actors dissenting with the executive increased steadily.

Without a demonstrated capacity to repress elite domestic political actors, the
executive elected to mitigate domestic political pressure through an alteration in policies
rather than face the prospect of deselection or more plausibly reduced political clout. The
accommodation that Mao reached with his constituents after the fifth Winter-Spring
Offensive allowed the PVA to create defensive positions and end hasty efforts to regain
the military initiative. Additionally, military modernization and training was also

undertaken to enhance battlefield effectiveness (J ecks 1982, 27). The cost-benefit

 

involved in the conflict is ‘whether to put politics or military affairs first,’ but it is not accurate to claim that
the opposition has promoted a ‘purely military viewpoint.’. . .The opposition is accused of seeking to
“usurp military power,” plotting a coup d’état and even seeking to restore capitalism. . .They have failed to
give sufficient obsequious credit to Mao, and most of the principal critics have even ’slandered’ his works.”
(Powell 1968, 257).

’7 “Mao was compelled to retrench strategy from his preferred policy, while fighting a rearng action for
‘people’s war’. ‘These, changes, in both political and military strategy, were not necessarily wrenching
choices for China. In military strategy there was already Mao’s state opposition to mechanicalism which
urged pragmatic reactions to the changing nature of war. Under this broad guiding principle, and while
maintaining certain elements of a people’s war, many changes could be made with relative ease.”’ (Segal
I985, 110).

64

calculation of domestic political actors improved as the attritional strategy employed
reduced the degradation of the PVA.

Still as the open-ended attritional struggle continued many domestic political
actors began to ascribe less utility to current war policies than did the executive. Efforts
by officially non-combatant states to reach a cease-fire agreement were rebuffed by the
Chinese in December of 1950 as were the January 1951 attempts by the UN (Zhu 2001,
28-34).58 Even with a less aggressive military strategy, the war still meant enormous
opportunity costs for the economy and other expenses that put a plurality of domestic
political actors at odds with the executive.

Mao appeared to garner more utility form the prolonged fighting while talking
strategy coinciding with the attritional phase of the war (Zhu 2001, 97). Elites in the
CCP were not so sanguine regarding the implications of an open-ended commitment in
Korea. In fact domestic political actors were almost stunned by the blasé discounting of
the material costs of fighting (West 1992, 401).59 Political pressure for the broaching a
negotiated settlement intensified as the indefinite length of the war resulted in a heavy
time discount being applied to the realization of benefits.

While in retrospect the People’s Republic likely did not face a realistic chance of
being demolished by internal forces, domestic political actors feared for the stability of
the regime.60 As early as 1951 American intelligence reports indicated that some

dissenters in the less ideological wing of the CCP would be receptive to peace overtures

 

5’ Mao even in 1952 found the war to be useful in continuing to mobilize the people, consolidate the
regime, and to train the military. (Mott and Kim 2006, 128).

’9 “Chinese doubts among leaders and soldiers alike about the accuracy of portraying the United States as a
‘paper tiger’ may have occurred already by the spring of 1951. . .Over the next thirty years real differences
over foreign and military policy during the Korean War began to surface and led to bitter and tragic
struggles within the Chinese leadership.” (West 1992, 401).

6° The regimes consolidation and fear of the people’s unwillingness to further mobilize as concerns to the
CCP although perhaps not a dire threats to the People’s Republic (Zhang 1995, 223).

65

(Foote 1990, 72). The enormous material and human costs borne by China had
precipitated a debate within the upper reaches of the CCP and political pressure built to

engage in peace negotiations (F oote 1990, 71-72).61

Figure 3: Distribution of Elite Opinion in China

. . After Defeat of the Fifth
US Drive on Yalu River Win fer-Spring Offensive
Executive’s policy Distribution of Distribution of
preferences/ Domestic Political Domestic Political
current war policies Actors’ preferences Actors’ preferences
\

Executive’s policy
preferences/
current war policies

 

 

 

 

 

 

: > ¢ 5
Less favorable More favorable Less favorable More favorable
relative utility relative utility relative utility relative utility

The costs of mobilization for the war were ab'solutely massive. However, the
level of mobilization was necessary to sustain the war effort and maintain the stalemate in

the Korean Peninsula. In order to relieve the burdens placed upon the People’s Republic

 

6' “. . .the war in Korea had created controversy in the PRC. The decision to intervene in the fighting had

been difficult for the Chinese leadership to take, and the financial and personal toll of the fighting in 1951
had been considerable. . .the British counselor at the embassy in Peking, had been informed in the strictest
confidence by an intelligence source that (Primer) Chou En-lai had stated in an address that the Chinese
had had to take part in ceasefire negotiations because of the adverse effects of the war on the national
economy. Chou [Zhou Enlai] reported that the Korean War “had already cost China more than the whole
of the Japanese war had done.” Moreover, signals that the Chinese had sent via the Indian Charge

d’affaires in Peking indicated that there was a genuine opportunity to conclude an agreement at this time.”
(Foote 1990, 71-77).

66

a settlement had to be obtained, which would require a change in issue-goods negotiation
position. As with Japan the level of mobilized resources was not the point of debate, it
was how to reach the best settlement possible and thereby reduce the resource
commitment to the war.

Armistice negotiations were begun in July 1951, but did not progress rapidly.
Evidence of Mao’s lack of interest in moving the conflict to a conclusion was indicated in
the opening negotiation offers (Hamrin 1994, 98).62 The issue-goods demanded included
recognition of the PRC, admission into the United Nations with the associated Security
Council seat along with American disengagement from Taiwan.63 These offers received
a very chilly response and the war of attrition continued.

In December 1951 the Chinese dropped their demands for a resolution of
tangential issues not directly related to the war. Negotiations offered a way to enhance
the probability of achieving the ante bellum issue-goods at stake as well as reduce the
time until these benefits would be realized. In fact the basic framework for the division
of the peninsula was arrived at fairly quickly (Zhu 2001, 77-79).

The principal issue that suspended a settlement was the question of repatriation of
prisoners of war (POW). A substantial minority of captured communist soldiers did not

want to return to the People’s Republic and North Korea, this served as a major source of

 

‘2 “Certainly, the prolonging of war into the 19505 contributed to internal repression, with serious
consequences for stability and regime legitimacy later in the decade. Some foreign affairs officials and
experts cautiously and quietly have criticized Mao for pursuing the Korean offensive beyond initial gains,
prolonging the conflict and escalating the economic and social costs. Many younger Chinese diplomats
privately view the decision to enter the war itself as a huge mistake, a result of ignorance of the outside
world and a black and white ‘peasant mentality’ shared by Mao and other generals running the country at
the time. According to this view, more educated and experienced ‘moderates’ like Zhou Enlai, who were
charged with executing foreign affairs, consistently lost out in the policy debates of the period.” (Hamrin
1994, 98)

63 In response Secretary of State Marshall declared that Taiwan would never be allowed to fall into
communist hands. (Foote 1990, 35).

67

embarrassment for both regimes. Yet the issue was not a substantial source of utility to
most domestic political actors (Xia 2006, 65-70). Especially when balanced against the
time discount until these benefits would be realized and the costs likely to be incurred
through further fighting. Disagreements between elites in the CCP and the executive
with respect to issue-goods negotiation position were the principal source of political
pressure.

While the executive was not in immediate danger of deselection (Gurtov and .
Hwang 1980, 60), Mao did fear that his political authority might be reduced (Sullivan
1986-87, 616). These tensions were reflected in the party discipline campaigns. The
ongoing Korean War was associated with purges of lower-level members of the party.64
A number of scholars have noted a connection between the ongoing Korean War and the
severity of the party disciplinary campaigns (Hamrin 1994, 98; Wint 1960, 63; Uhalley
1988, 99). Purges in the 19503 were mainly an instrument to consolidate the communist
regime’s hold on power. However repression was also used against lower echelons of the
party in part to demonstrate the executive’s authority.

As Stegal notes (1985, 106), it is difficult to pinpoint exact instances of the policy
changes during the Korean War that were foisted upon the executive. Still there is
evidence that political pressure was applied to Mao in order to accommodate the

preferences of domestic political actors. Even though deselection was not a serious

 

6’ “However, the linkage between external threat and severity tended to be borne out by the relative
harshness of the movements conducted during the fiercest fighting—the intense struggles of the sanfan
campaign and the high purge rate apparently set for Party rectification in this period. And although there
were clearly other, perhaps more important factors, in 1953 with the war stalemated and a negotiated
settlement near, the New Three Anti Campaign was markedly milder than its immediate predecessors.”
(Teiwes 1993, 127).

68

threat, in that Mao was so strongly identified with the regime, the sanction implied by
political pressure was the comparative decline in Mao’s political standing.

Even as strategy was scaled back from a more aggressive and ideological tact,
these efforts appeared to buy Mao time to continue to prosecute the Korean War in
accordance with his issue-goods preferences. The growing disquiet over the future
security of the People’s Republic amongst domestic political actors began to force Mao
away from his preferred policies.

The executive eventually opened serious negotiations acquiescing to the
preferences of many CCP elites. Demands that negotiations progress more quickly
toward a resolution meant that Mao’s preferences for tangential issue-goods were
dropped fi'om armistice talks as well as the decision to relinquish the position for
involuntary repatriation. While the political pressure being applied to Mao was
considerably less than that placed on the Japanese executives, the Chinese case indicates
more responsiveness to elite disquiet over war policies.

Had battles revealed the relative balance of capabilities and resolve in accordance
with an objective cost-benefit assessment likely the Korean War would have been fought
differently. The executive’s preferences in terms of war policies were adjusted in
compliance with the demands of his constituents. Even in the face of heavy losses, the
executive did not make an effort to alter military strategy.

Effectively the objective criterion of casualties was not the catalyst for a change
away from people’s war, instead it was the demands placed upon the executive by
domestic political actors that appeared to drive changes in strategy. This does lend some

merit to the Gartner hypothesis that key factors such as troop losses appeared to drive an

69

alteration in tactics. But it was only under duress that Mao moved away from his
preferred strategy.

Changes in war policy reflected a median course between disparate factions
within the Beijing Politburo or they reflected Mao accepting advisement of CCP officials.
Yet Mao appeared to derive much greater utility than did most domestic political actors
in the regime. The comparatively isolated opinions of the executive did not appear to
represent an aggregation of the collective elite assessments. As evidenced by the
disciplinary campaigns of the communist party serving as a proxy struggle between Mao
and some of his constituents.

The selection of the Korean War was accepted by elites in the CCP as a costly but
necessary endeavor. However, having secured the immediate objectives of maintaining a
buffer between an American ally and mainland China support moved to a resolution of
the conflict and critiques of how the war was conducted. Instead of the executive
gambling to redeem an absolute victory, most of the conflict was spent as an attritional
struggle over issues peripheral to the question of North Korean sovereignty. In a regime
with a strong capacity to suppress popular dissent but a questionable ability of the
executive to repress elites, no attempts to take more hazardous risks for a clear victory
were made.

The situation that faced the People’s Republic of China during the Korean War
shares the characteristics of the Japanese executives in that a limited ability to repress
elites was demonstrated. Although, the threat posed by the war in Korea was not nearly
as serious to the regime’s survival as was World War II to Imperial Japan. Consequently

when Mao refused to accept a reduced settlement during the extended negotiation

70

process, his hesitation did not inspire an imminent threat of deselection. But as domestic
political actors ascribed a reduced amount of utility to the war the resultant political

pressure helped facilitate a peace settlement.

Iran: Iran-Iraq War

The shah was deposed in 1979 by a collection dissident groups, including leftists,
constitutionalists, ethnic minorities, and Islamic parties. While the Ayatollah Khomeini
set the tone for the revolution upon his return from exile, a cacophony of groups
compelled the monarch’s abdication. In the midst of creating a new regime war was
foisted upon the young republic.

Absent a unifying antagonist in the shah, competition for power amongst
competing revolutionary groups emerged (Hiro 1985, 103405).“ In the wake of the
1979 revolution a number of political institutions were formed, the power of the branches
of government was largely determined by the personalities in these positions rather than
the de jure authority ascribed to them (Moin 1988, 192-193). At the onset of the war
with Iraq government officials, politically engaged Shiite clerics, the military, and leaders
of other political parties who helped depose the monarch constituted the state’s domestic
political actors.“ A fluid political situation offered a disparate collection of domestic
political actors capable of potentially supplanting the de facto executive the Ayatollah

Khomeini."

 

’5 Hiro (1985, 103-105) lists the Fedai Khalq and the Tudeh as Marxist-Leninist parties as being
particularly influential. Quickly after the Shah’s abdication these other political movements clashed with
the Ayatollah demanding a more secular regime.

‘6 Nasr helps draw the distinction between the more politically passive Shiite spiritual leaders and
Khomeini’s highly political movement (2006).

67 Constitutional institutions included the “Assembly of Experts” directly elected theologians, the “Council
of Leadership” a position(s) essentially made for Khomeini as a supreme leader and effective veto over the

71

The most important cleavage in Tehran’s politics was between Khomeini’s
Islamic Revolutionary Party (IRP) and the moderate-left embodied by President Bani-
Sadr.68 The president was substantially less powerful than Khomeini, yet Bani-Sadr still
enjoyed control over a limited portfolio of policies including the military (Reisner 1989,
66). Bani-Sadr also constituted one of the most plausible candidate 8 to supplant the
ayatollah.

Observing the disordered Iranian polity following the Shah’s abdication Saddam
Hussein decided upon a unilateral adjustment of the 1975 border agreement. The Iraqi
executive thought the time right to exert his influence in the region and mitigate the threat
of a Shiite revolutionary movement on his border.69 Instead Hussein engendered the very
risk he sought to avoid (Walt 1992). The Iraqi assault was to give Khomeini an
opportunity to quash domestic critics in the name of nationalism and to proselytize the
Shiite revolution amongst neighboring co-sectarians (Chubin and Tripp 1988, 38).

While the Iraqi invasion would not be remembered as one of history’s great
offensive campaigns, in September 1980 it was a major crisis for Tehran. Bani-Sadr’s de
jure authority over the security policy was quickly challenged by a divided command

structure.70 A parallel military system came into place as the political militia of the IRP,

 

presidency and determining issues of war and peace. A judiciary “Council of Guardians” that exercise
Judicial review over the assembly Majlis. The Majlis elects the prime minister and a president is directly
elected. (Main 1988, 192-193).

6’ Moin identifies three factions amongst the Islamic government the traditionalists, reformists, and
radicals. Traditionalists of whom Khomeini was the leader was the dominant faction (1988, 193).

69 It has been observed that Saddam believed he was securing his regime against the threat posed by the
emergence of a Shiite theocratic state on his border. In effect he had to limit this threat to his rule (Razi
1988, 699).

7° The purges of military officers had stripped away professional expertise and many survivors of these
anti-monarchist crusades had resigned their commissions or emigrated. Moreover the Shah was a major
consumer of American military equipment and advisers, following the hostage crisis the supply of both
were shut off markedly reducing the number of serviceable weapons in the Iranian arsenal. What a few
years before had been one of the most potent militaries in the developing world, was now greatly reduced
in capabilities (Zabih 1988, 115-126).

72

Islamic Revolutionary Guards (IRG) or Pasdaran, evolved into something more
representative of a formal army." Both forces participated in a vigorous Iranian defense
that stalemated the initial Iraqi invasion. Hussein reappraised the costs and benefits of
further fighting and with little evidence of an Arab uprising in Iran or collapse of
Khomeini’s authority through political infighting, Baghdad signaled its willingness for an
ante bellum settlement (Hiro 1984, 6).

Hussein’s acquiescence to peace talks was not to be reciprocated. The Iranian
executive joined in calls for a religious crusade against the secular Baathist regime
(Chupin and Trip 1988, 73).72 Khomeini and his supporting domestic political actors
derived a great deal of utility from an ongoing war. The war with Iraq permitted the
alienation of alternative political parties aided in part by their links to Baghdad, Moscow,
and other foreign regimes. The conflict also helped distract the military from internal
politics, and rationalized an increased role for the IRG in political affairs (Hiro 1985,
237; Chubin and Tripp 1988, 40-41).

Seeking to consolidate his authority Khomeini isolated the president removing
him as a potential rival executive. As the war continued into 1981 Bani-Sadr was
politically outrnaneuvered. His fellow sympathetic domestic political actors in the new
regime were “decimated” leaving a marginalized president struggling to maintain his
policy portfolio. The elimination of this prominent domestic political actor was
accomplished when Bani-Sadr was accused of having contacts with the CIA and deposed

in 1982 (Rouleau 1981).

 

By one accountant 30% of the officer corps had been removed from office in 1980 (Halliday 1980, 3-5).

7' Although these were lightly armed by conventional military standards and a lose command and control
structure (Zabih 1988, 209-211).

72 The Speaker of the Majlis (parliament), Rafsanjani, indicated in March 1981 that Iranian objectives now
included the removal of Hussein fi'om power (Hiro 1984, 6).

73

Intense violence was ignited by the ousting of the president (Mackey 1996, 301-
307). But the IRP and Islamic hardliners prevailed in these battles eventually defeating
the Marxist Tudeh party and the Islamic-socialist Mujahedeen party politically as well as
in urban combat. The IRP consolidated its position into a near monopoly on power as the
last of the pre-revolutionary parties was banned in May 1983 (Hiro 1985, 251). Having
altered the body of domestic political actors capable of challenging his authority,
Khomeini continued to pursue his preferred war policies.

At this point the Iranian regime adopted characteristics more common to the other
cases studied here. By eliminating rival claimants to his position a more consolidated
political system emerged in Tehran. Now the ideological focus of the regime was
determined and the principal interplay now lay between competing factions in the Islamic
Republic.

Additionally the initial war aims of the defense and reconquest of Iranian territory
had been a near universally accepted issue-goods position at the onset of the war despite
the fragmentation of the state’s politics. The winnowing of domestic political actors to
ones more ideologically congruent with the executive meant that the new war aims of
destroying the Baathist regime retained more appeal than would have been the case had
the original composition of Tehran’s politics endured.

In pursuit of defeating the Baathist regime, both the IRG and the Iranian army
used an attritional strategy as combined arms fighting using mechanized assault and air
assets could not be realized due to a shortage of serviceable equipment (O’Ballance 1988,
68-69; Chubin and Tripp 1988, 42-43). While these tactics offered no serious

improvement upon the strategy of the Western Front of World War 1, human wave

74

assaults were able to eject the Iraqi army from Iranian territory in the first half of 1982.

These same tactics were then employed to topple the Baghdad regime.

Figure 4: Distribution of Elite Opinion in Iran

After military purges and

Iran lnvaswn elimination of Bani Sadr

Executive’s policy

preferences/ Distribution of Distribution of
current war policies Domestic Political Domestic Political
\ Actors’ preferences Actors’ preferences

Executive’s policy
preferences/ —+

current war policies

 

 

 

 

 

LV : 4 5

Less favorable More favorable

. . . , , , Less favorable More favorable
relative utility relative utility

relative utility relative utility

Despite success in reclaiming Iranian territory, the professional military was less
optimistic regarding the likelihood of defeating Hussein’s army. Having little sympathy
for the executive’s new issue-good objectives an attempt to deselect the executive was
manifested (Milani 1994, 186; Hiro 1985, 219).73 Before the coup could be executed the
plot was discovered resulting in extensive purges and executions of military officers and
other conspirators conducted from April to August 1982 (Zahbi 1988, 129). These

actions cowed the military as a political force and further altered the composition of

 

7’ While the conspiracy’s principal impetus came from military officers, it was joined by at least one
civilian official, the former foreign minister during the hostage crisis (Milani 1994, 186; Hiro 1985, 219).

75

domestic political actors in the state. With a more sympathetic and somewhat smaller
class of constituents, in July 1982 Khomeini took the war to Iraq.

A number of offensives were launched against the southern part Iraq attempting to
cut Basra off from Baghdad and thereby spark a Shiite revolt. Baghdad’s armies
countered these attacks by using superior weaponry, maneuverability, and prepared
positions. Even when the Tehran’s soldiers captured ground, holding it when faced with
vigorous counterattacks proved difficult. This was especially true of the ideologically
charged but poorly trained IRG troops whose disorganized infantry assaults struggled to
press their hard won gains into major breakthroughs. The Iranians also faced logistical
difficulties supplying troops at the front, further reducing the likelihood of exploiting
breaks in the Iraqi lines (Chubin 1989, 7-9).

Nationalism and gritty determination had ejected the Iraqis from Iran, but the
effort to destroy the Baathist regime was shown to be difficult endeavor. Whereas a more
accelerated pace might have had greater success in breaking Iraqi morale the massive
attritional style attacks followed by long lulls as preparations were made for another

campaign, stalled any momentum.“

 

7" “If Iran’s military successes between 1982 and 1986 were ephemeral and costly, with long gaps between
major offensives from 1984-86, the problem stemmed as much from deficiencies in strategy as from
logistics. Alternating between frontal offensives and attrition along the length of the hamlet. . .between
enthusiasm for the daring of the revolutionary guard and the more sober appraisals of the professional
military, Iran’s leaders were unable to frame a strategy that tied their war aims—the overthrow of the
enemy—to their military capabilities, which in terms of equipment dwindled with each offensive. To
achieve their war aims. . .Iran needed either to defeat the enemy’s forces decisively, or to capture a major
strategic asset, thus precipitating their surrender [for example, the southern city of Basra, which was
predominantly Shii]. The problem was that Iraq’s forces would not venture out into the field to fight and
risk defeat which the capture of Basra or Baghdad remained increasingly difficult because of their
redundant defense lines.” (Chubin 1989, 4).

76

Whereas western observers tended to believe that Iran with triple the population
of Iraq might be able to wear down the Baathist regime.” This didacticism did not
sufficiently grasp Iraqi adjustments to the war of attrition. Despite his reputation as
anything other than a humanitarian, Hussein sought to minimize Iraqi fatalities for fear of
what heavy casualties could mean for the regime’s stability (Chubin and Tripp 1988, 58-
59). In addition to minimizing human losses the Iraqi executive also ordered a series of
punishment style attacks against Iranian oil production and surface to surface missile
attacks against cities (Chubin and Tripp 1988, 61-63). The bombardment of Iranian
urban centers did not seriously undermine the war effort, but the threat to oil exports was
a serious matter.76

The diplomatic isolation of Tehran following the hostage crisis meant sanctions
and imposed autarky. When coupled with market disruptions from the ongoing war Iran
faced a stagnant-shortage economy (Chubin and Tripp 1988, 123-138). Expensive state
funded subsidies were needed to bolster civilian morale, diverting funds from
investments in oil production. The war’s economic opportunity costs were especially
damaging to the regime in light of the IRP’s drive to reduce shah-era income inequities
(Karsh and Rautsi 1991, 173). Finally, the surreptitious purchase of weapons due to the
comparatively stringent arms embargo meant that it took Iran $3 to purchase $1 worth of

arms (Wenger and Anderson 1987, 26).

 

7’ The fallacies associated with this conventional wisdom it was Iranians who could not afford to accept
stalemate (Chubin 1989, 5).

76 While the price of oil initially spiked after the onset of hostilities it soon returned to a more moderate
level in the mid-1980s. Further, bereft of western expertise and facing mounting attacks by Iraq oil
production declined (Reissner 1989, 66-67). The war also required large infusions of cash to secure
military equipment, but the comparatively strict arms embargo imposed on Iran made weapons much more
expensive in that they were purchased surreptitiously. Eventually the Iranian armed forces were forced
away from their American arsenal toward Soviet weaponry with large expenditures of hard currency. (0’
Ballance 1988, 102-104).

77

More important than economic stagnation, the human costs of the war were
undermining support for the regime. In early 1984 the Iranian Val Fajr 6 Offensive was
spearheaded about thirty miles north of Basra in an attempt to help foment a Shiite revolt.
This initial assault was followed by several follow on attacks with little to show for the
sacrifice other than a harbinger of more slaughter to come. By March 1984 the massed
infantry attacks against Iraqi defenses, along with all previous fighting, had cost Tehran
170,000 killed (Hiro 1991, 105).

While the promise of paradise was a potent incentive in inspiring soldiers, the
death of thousands was a very dear sacrifice to proselytize Khomeini’s revolution.
Increasing public discontent found some resonance with Iranian domestic political actors,
as illustrated in the following passage.

Popular dissatisfaction echoed within the councils of the Iranian revolutionary regime as early as
mid-1982, when a loose coalition of military and political figures began to question the logic of
taking the war into Iraq on the grounds of its human, material and political costs. As national
spirits fell, such skeptical voices became increasingly influential. Nevertheless, it took nearly five
years of growing dissatisfaction before the authorities could fully reconcile themselves to the
futility of the conflict. Karsh and Rautsi 1991, 173

Whereas discontent with war policies had previously come from elites outside of the
Islamic revolutionary movement, intense dissatisfaction with the war now emerged from
elites sympathetic to the theocratic regime. Despite Khomeini’s strong position afier the
destruction of the leftist political parties, removal of president Bani-Sadr, and eradication
of a military conspiracy, the factionalization of the Islamic Revolution was well
underway (Reissner 1989, 69).”

Costs in the form of regime stability were paramount as the material and human

expenditures mounted (Moin 1988, 196; Karsh and Rautsi 1991, 173). Many elites

 

77 “After having stamped out any conceivable remaining traces of opposition from outside the ruling elite,
cleavages soon opened between different factions within the leadership about the future course of the
revolution.” (Reissner 1989, 69).

78

assessed a declining likelihood of achieving the overthrow of the Baghdad regime as well
as a greater anticipated time until these benefits could be realized. Political pressure
mounting since the decision to take the war to Iraq took form as a lobbying effort for an
alteration in issue-goods position to one that would be readily accepted by Iraq (Chubin
and Tripp 1988, 82).78

Political pressure had emerged in the Iranian context earlier in the war as the
leftist, secularist, the officer corps, and Islamic domestic political actors assessed the
costs and benefits of destroying the Baghdad regime differently. But after Khomeini
eliminated opposing domestic political actors the level of political pressure subsided.
Even in the context of smaller class of domestic political actors, political pressure would
return as the perceived utility of current war policies was assessed negatively.

Facing emergent political pressure and without a developed capacity to intimidate
his opponents within the IRP, the executive was equipped with few options to enhance
perceptions of the war’s utility. Being so heavily invested in demands for an end to the
Baghdad regime meant a change in issue-goods position would be deeply effacing for
Khomeini (Razi 1988, 706). Similar to the Japanese and Chinese cases, the executive’s
reticence to alter issue-goods position to one which had a plausible chance of being
accepted meant that the war distended onward. Little consolation for political pressure

was offered in the form of a revised bargaining position.

 

7’ “On 4 February 1985, twenty-six personalities associated with the Islamic republic criticized the policies
of the leadership in a manifesto that was circulated, referring to the human and material losses incurred in
the war without explicit criticism of it. On 18 March, Bazargan [interim prime minister immediately
following the revolution] and sixty supporters cabled the UN Secretary General about the war, labeling it
un-Islamic and illegal since mid-1982 [when Iran recovered its territory]. There were signs of some
clerical dissatisfaction as well. Ayatollah Hassan Qomi, in the first authoritative criticism of the war, on 5
March described it as ‘religiously unlawful’. This was said to have been echoed by Ayatollahs
Golpayegani and Morteza Haeri. It was also said that Ayatollahs Meshkini, Azeri Qomi, and Tabatabai, as
well as the Prosecutor-General Sane’l, at various times voiced criticism of the continuation of the war.

79

Likewise little innovation was registered in strategies used at the front. In large
part this reflected the inability of the Iranian military to compile the resources needed to
alter military tactics. While basic rocketry and air strikes against Iraqi shipping and oil
facilities had been accomplished, even these assaults had been limited in scope and could
not be sustained by the diminishing capital stock of advanced aircraft (Cordesman 1987,
213). Without anything approaching an effective punishment strategy that could achieve
the executive’s sought after issue-goods, the gruesome attritional struggle continued.

In Februaryl986 a massive assault was undertaken against the Fao Peninsula in
Iraq, a piece of territory containing a large percentage of Iraq’s oil reserves. The area
was captured, but the Iranian military absorbed a huge mnnber of fatalities from both
conventional and chemical weapons for the effort (O’Ballance 1988, 179).79 As with the
Japan’s defeats at Midway and the failure of Mao’s fifth Winter-Spring offensive,
battlefield events refashioned the impressions of Iran’s elites. Though successful the
heavy costs of the campaign altered perceptions of the war’s utility for a broader cohort
of domestic political actors (Hoogland 1989, 6).80

The waxing level of political pressure amongst Islamic Republic elites was
responded to by Khomeini’s attempt to break the stalemate, rather than countenance
reduced mobilization. The expensive tactics employed by Iran throughout years of
conflict had produced some successes, at one point nearly severing the road link between
Basra and Baghdad (O’Ballance 1988, 143). Using these pyrrhic victories as a template,
a “final offensive” was formulated by the executive to terminate the war (Hooglund 1989,

6). The attempt to fundamentally alter the dynamic of the war did not occur. Instead of

 

79 An Iraqi assault would reclaim the peninsula in April 1988.
3° “The Faw victory ended up persuading even more leaders of the need to find a non-military solution to
the war.”’ (Hoogland 1989, 6).

80

one, a series of offensives were launched throughout late-1986 until early-1987. None of
which were able to break Iraqi resistance or foment a Shiite rebellion.

The military situation after the failed offensives of 1986-87 resulted in exhaustion
of the Iranian populace. Unrest was threatening the regime and more coercive methods
were needed to quiet dissenters (Reissner 1989, 68-69). Domestic political actors
accorded greater costs to the further prosecution of the war as the potential of internal
revolt grew.81 The likelihood of toppling Hussein seemed remote and the resources
mobilized for the effort appeared to be undermining the regime’s stability. Ongoing
mobilization was increasingly difficult as casualties and economic deprivations of war
were reducing the inspirational advantage that Iranians enjoyed over their opponent
(Chubin 1989, 7-8).

The success that the executive enjoyed in reducing the size and diversity of
interests of domestic political actors earlier in the war helped maintain support for war
policies. It was only when the endangerment of the regime became palpable that political
pressure grew especially great. As elite discontent with the war increased it was the
political authority of Khomeini that was diminished, rather than deselection (Chubin
1989, 20-23). The de facto power of the executive was degraded and more pragmatic
military approaches were undertaken, such as the Rafsanjani facilitation of an arms deal

with the US (Razi 1999, 704).

 

3’ “From 1986 onwards, it became clear to at least some members of the Iranian leadership that the balance
between the effects of revolutionary mobilization for the war and the war’s drain of resources was
beginning to shift to Iran’s disadvantage. This led to a last dramatic effort to reach victory in the land
war—the offensives of December 1986 to February 1987. When the military breakthrough proved elusive,
Iran’s situation deteriorated rapidly.” (Reissner 1989, 67).

81

Seeking to end the war, Hussein launched a series of strikes in the spring of 1988
that pushed the Iranian armed forces out of Iraq. These defeats demonstrated the small
and diminishing probability of coercing regime change upon Baghdad. Battlefield
outcomes resulted in a growing rift between a majority of domestic political actors’ and
the executive’s assessments of current war policies. It was then that domestic political
pressure manifested itself in the effective removal of control over military policy from

Khomeini to other elites in the Islamic Republic.

During this same year popular exhaustion with the war and its attendant privations had increased
dramatically and had begun to affect the performance of conscripts at the front. Discussions after
July 1987 focused more and more on the domestic implications of the war. The unexpected
military reverses in the spring of 1988 caused further anxiety. By June, a majority of top leaders,
including President Ali Khamenehi, Prime Minister Mir Hosain Musavi, and Speaker of the Majlis
Ali Akbar Hashemi Rafsanjani, had become convinced that continuation of the war posed a
serious threat to the stability, even survival, of the Islamic Republic. Other leaders, including
Khomeini, were not ready to concede that the situation was quite so grave. The shoot down
provided a dramatic incident that proponents of the ceasefire seized upon to argue the case for
accepting Resolution 598. Future developments were entirely unpredictable if US involvement in
the war expanded, they argued to their colleagues, convincing them that a UN supervised cessation
of hostilities was the best course to follow so that Iran could put aside and get moving again on its
internal revolutionary agenda. Tehran notified Perez de Cuellar that it accepted the Security
Council’s document without reservations. Subsequently, Iran’s leaders persuaded Khomeini
himself to sanction their decision publicly. Hooglund 1989, 8

Political pressure emerged as Khomeini was compelled to reverse his preferred war
policies in the face of mounting political pressure. Already witnessing a reduction in his
political position the executive acquiesced to a marked change in issue-goods negotiation
position.

The negative utility of further fighting compelled Khomeini to settle for ante
bellum status quo offer that provided immediate termination of hostilities. This was a
remarkable move for an executive so personally tied to the cause of exporting the Shiite

revolution (Reissner 1989, 69).82 Political pressure had compelled an executive to part

 

’2 “Khomeini’s decision to accept a cease-fire represented a truly dramatic policy reversal, which, at least in
terms of the revolution’s own rhetoric, posed a serious risk for the revolution itself. Yet the separation of

82

with his preferences and seek a quick peace, moving the issue-goods position from one of
high issue-goods salience to one of much reduced personal utility.

While Khomeini demonstrated an ability to eliminate dissident domestic political
actors, this capacity was not as great or was not as actively employed once the Islamic
movement had assumed control over the government. The consolidation of IRP into the
de facto government also limited the diversity of opinions regarding war policies.
Resembling more closely the CCP rather than Tokyo’s more diffused oligarchy.

In terms of the intensity of political pressure Iran in its war with Iraq faced the
same threat to the govemment’s existence only when the exhaustion of the population
was undermining the legitimacy of the Islamic Republic. With a more developed ability
to repress elites Khomeini was able to endorse his preferred war policies for most of the
war. In contrast to the Japanese executives who were deselected after refilsing to
acquiesce to demands for changes in war policy. Even in the face of political pressure
the executive was able to eliminate dissenting voices.

However when the danger to the regime grew the political pressure mounting for
a resolution to the conflict could not be ignored. Whereas Mao accommodated political
pressure through changes in strategy and later issue-goods position, Khomeini did not.
Instead the Iranian executive escaped deselection, but lost much of his political authority
by the end of the war.

If battlefield outcomes were appraised objectively, then there would have been no
need to eliminate military officers and with them their professional expertise. Had the

results from the battlefield been assessed in a unitary fashion then policies would have

 

the revolution from the war, long presented as inextricably intertwined, was now achieved with remarkable
ease under the slogan of ‘saving the revolution’”. (Reissner 1989, 69).

83

been undertaken that reflected this shared understanding of current war policies. Instead
dissent with the executive’s choices manifested itself in lobbying and the eventual
political degradation of Khomeini’s authority.

The alternative hypothesis that the state would pursue a more aggressive strategy
in order to achieve victory is at least partially supported in the above analysis. Khomeini
launched massive assaults with tremendous human costs in an attempt to win the conflict.
While it is unclear if the executive anticipated punishment at the end of the conflict, there
was a palpable sense of political pressure. Troublesome for Goeman’s risk acceptance
hypothesis is that while the assault was launched to win the war a status quo resolution
was always possible should the executive accept it. Likely he would have survived, and
in fact did, with a tie instead of a win. Moreover it is likely that the massive offensives at
the latter part of the war were the cause of political pressure rather than a device to
mitigate elite concerns. The Iranian case does not appear to be a ringing endorsement of
the risk acceptance theory.

The election of military strategy was apparently also not strictly tied to a
calculation as to losses. In fact the attritional infantry attacks varied little throughout the
war. Whereas some punishment style assaults upon Iraqi oil production was undertaken
this was constrained by the lack of resources. The enormously costly infantry assaults
were only staunched by the failure of the public to mobilize rather than by Tehran’s
volition, suggesting that it was not changes in the rate of combat losses which compelled
the alteration of strategy.

The Iranian case offers an executive with a more intermediate ability to repress

elites. Little accommodation is made to reform the manner the way the war is fought,

84

with mobilization, military strategy, and issue-goods negotiation position changing little
in the face of political pressure. However, as the threat to the regime grew so did elite

dissent and the executive eventually bowed to political pressure.

North Vietnam: Vietnam War

The Vietnamese Worker’s Party (Lao Dong) presided over a land that was
underdeveloped but highly mobilized for the primordial cause of Vietnamese
nationalism. The ejection of the latest foreign power and reunification of the country was
a consuming objective to the leadership of the Democratic Republic of Vietnam (DRV).
While the ends of the war were not in question, the means of achieving reunification were
a subject of intense debate.

Ho Chi Minh was enormously influential and without his dominating presence
Hanoi elites began internal struggles over policy.83 As Ho’s health deteriorated in the
mid-19608 a collective system of leadership emerged in North Vietnam. The relevant
domestic political actors of this party-state were the Politburo members and their deputies
as well as prominent military leaders. Within this group all politburo members were
equal with Le Duan, the new executive, being more equal than the others (Pike 1989).“

The cleavages along domestic political actors fissured revolved around the

question of how to seek reunification with the South while simultaneously creating

 

’3 “The decline of Ho’s influence to a large extent exacerbated the leadership struggles and policy acrimony
in the Hanoi Politburo, despite designating Le Duan to take up the mantle of the executive.” (Thai 1985,
52).

8’ At least one source disputes this characterization: “The thesis set forth by some western scholars. .
.according to which Ho had institute a model of collective leadership, corresponded more to the image that
the Lao Dong wished to project within the country and abroad rather than the actual reality.” (Thai 1985,
52).

85

socialism in the North. This division also heavily correlated with the relative affinity
with which domestic political actors viewed the two major communist powers.85

The Moscow oriented Politburo members tended to prioritize the conflict in the
South somewhat lower than the construction of a socialist (industrial) economy in the
North.86 The pro-Soviet domestic political actors supported a protracted warfare through
guerrilla tactics in order to forward reunification.87 Generally this was taken to mean that
large communist units would avoid contact with the Americans and prolong the conflict
in order to sap their antagonists resolve. Accordingly the utility calculation of this group
generally reflected a less precipitous time discount of future benefits. Also these elites
considered the costs of direct military confrontation with the United States as being more
onerous than did their Opponents in the Politburo. ‘

The pro-Beijing faction embraced a multi-staged path for the conduct

revolutionary warfare (Powell 1968). Guerrilla campaigns were to be followed by

progressively larger combat units fighting a more conventional style of warfare.88 The

 

’5 “Although there were significant differences between the top DRV leaders on the question of policy
priorities during the 1968-71 period, the political debate on this issue cannot be regarded as a manifestation
of a power struggle between organized opinion groups or personnel factions among the North Vietnamese
leaders.” (Ang 2002, 41). Although, strong evidence for Moscow vs. Beijing orientation is also found
(Thai 1985, 8).

’6 “Intraparty tendencies reportedly can be divided into three categories: pro-Soviet, pro-Chinese, and a
selective combination of pro-Soviet and pro-Chinese policies. The pro-Soviet group is led primarily by Vo
Nguyen Giap and includes Pham Van Dong and Pham Hung. It is confronted by the larger pro-Chinese
faction reportedly led by Truong Chinh and including Nguyen Duy Trinh, Nguyen Chi Thanh and Le Duc
Tho. The third group, which adheres to a position of ‘positive neutrality’ on party problems is headed by
Le Duan and apparently has the backing of Ho Chi Minh.” (Yearbook on International Communist Aflairs
1968, 639-640).

87 Moscow was known to periodically encourage a negotiated settlement with the Americans, besides their
intervention in the Geneva Settlement (a treaty creating the post-French division of the country), they were
encouraging talks as early as 1964. (Ang 2002, 83).

’8 “Embedded strongly into the fabric of communist ideology in North Vietnam was the Maoist perspective
on Marxism and the example of the Chinese Civil War. Without an industrial workforce to enshrine as the
source of revolution spawned an alternative perspective of what revolution should look like. Also
mirroring Beij ing’s experience in the Korean War, another fissure emerged between those that endorsed a

modernization of the military as opposed to those who supported a politically oriented revolutionary force.”
(Turley 1969, 882).

86

South-first faction advocated increased mobilization of resources in order to hasten the
end of the conflict accepting heavier losses associated with attacking major American
military formations. The dangers associated with confi'onting American force in South
Vietnam were generally considered acceptable sacrifices in order to ensure a quicker
realization of issue-good benefits.

The intervention of the US after the Tonkin Gulf incident and the beginning of the
Rolling Thunder bombing campaign in 1964 greatly increased the costs of reunification.
Faced with this shock to the utility calculation for prosecuting the war in the South, what
had otherwise been a comparatively amicable set of differences took on a substantially
different tone. Doctrinal debates were so acrimonious that they spilled onto the pages of
official state publications revealing tensions to the broader citizenry (Duiker 1996, 275-
276)

The precipitous decline of the Saigon regime’s military capacity resulted in the
May 1965 introduction of battalion sized US combat formations to protect airfields.
Reflexively the large-unit faction of the Politburo wanted to introduce North Vietnamese
Army (NVA) units to counter the American buildup.

General Giap, commander of the communist forces at Dien Bien Phu and leader
of the protracted war cohort of domestic political actors, thought this profoundly unwise.
By contrast the effective executive Le Duan contended that while the current policy of
supporting Viet Cong (VC) insurgency was comparatively inexpensive, the efforts to
destroy the Saigon regime appeared stalled. The extension of the time likely until the
realization of benefits was steeply discounted by the pro-Beij ing/large-unit faction and

the executive (Ang 2002, 117). The imperative for a quick victory was further fueled by

87

the deterioration of recruiting efforts and faltering morale of communists in the South as
the American military presence grew (Durkier 2000, 557; Durkier 1996, 287).

In order to facilitate a greater amount of mobilization and suppress dissent to his
war policies the executive initiated a limited campaign of repression. Domestic political
actors opposing the executive’s preferred war policies were subject to sanction despite

the “collective” nature of Hanoi’s politics.89 As illustrated in the following passage.

As a harbinger of things to come, three of Giap’s top aides were arrested in 1964 for being
‘revisionists.’ With Giap and opponents of direct military intervention in the south weakened,
purged or arrested, the Central Committee’s Twelfth Plenum in December 1965 Opted for a
strategy that entailed small-scale offensives, heavier causalities, a greater risk of escalation, set
piece battles requiring Soviet equipment, and larger amounts of aid that could not be provided by
China alone. Abuza 2001, 61

Greater mobilization was sought to hasten the conclusion of the conflict and reduce the
time discount upon the realization of issue-good benefits.

Having tamed opposition to his preferred policies, Le Duan ordered the
deployment of large NVA units into the northern part of the Republic of Vietnam (RVN)
with the intent of securing the passage of materiel support for the insurgency. Some of
the best units under Hanoi’s command were ordered to engage American formations on
the most advantageous terms possible.

As was predicted by Le Duan’s political opponents, the results were very
unfavorable.90 Baptism by American firepower had made the already skeptical General

Giap into an acrimonious critic of the executive. The costs associated with decimation of

 

’9 As was also the case in China, lower level members in the political hierarchy are often targeted in purges
in order to cow higher ranking officials. Other softer tactics were also employed with mixed effect to
revent “factionalism” in the Politburo. (Pike 1989, 121-122).

“Giap’s soldiers had their first significant contact with US troops at Plei Me [October 1965]. Then came
the battles o the la Drang [November 1965] as Giap’s men fought with elements of the American 1” Air
Cavalry Division (Air Mobile). Long convinced of the folly of the Le Duan/Nguyen Chi Thanh strategy,
Giap was now faced with firsthand evidence that his forces could not maintain face-to-face battles with
American forces. It was time to draw back. . .What was needed was further protracted war, to grind down
the enemy.” (Currey 1996, 81).

88

elite NVA units was especially disconcerting given the potential for a US invasion of the
North and further disruptions of the DRV’s economic development (Van Dyke 1972, 84).

Observing the results from the battlefield the North-first contingent of domestic
political actors assessed the utility of current war policies less favorably than did the
executive and his supporters. The application of political pressure to change military
strategy commenced. The imminent prospect of the deselection of Le Duan is not at all
clear, still the possibility of reduced political influence or leverage over opponents meant
that mobilization could not be assured of taking place.9|

Political pressure from these domestic political actors, which now included senior
officers operating in South Vietnam, took the form of lobbying for a reversion to guerrilla
warfare (Currey 1996, 275).92 Political pressure resulted in a shift in strategy toward
more synergy between the VC and the NVA and adoption of guerrilla tactics (Currey
1996, 276). The shift toward a lower intensity strategy was in no way final and in June
1966 plans were made to besiege the US Marine base at Khe Sahn.93 The ability of Le
Duan to make wholesale changes of the Politburo was beyond his faculties; still
repression was used as a means of cowing dissent.

In July 1966 official state publications declared the North-first/protracted war

Politburo members as being “pacifists” who were not sufficiently motivated for the cause

 

9' By most accounts Hanoi enjoyed a fair amount of stability in the upper reaches of power when compared
to other communist systems. (Turley 1980, 2-3).

92 “Giap’s views found a sympathetic response among some members of the southern military command.
One apparently high-ranking military leader in the south, writing under the name ‘Cuu Long’ and
presenting his views on liberation radio in the fall of 1966, took issue with the current big-unit high-cost
strategy. Claiming that large forces do not guarantee victory, be emphasized the need for more flexible
tactics in the South. In a second article published in a DRV military newspaper two months later, be
praised the role of guerrilla war in the revolution and bluntly complained that it was being underutilized in
south Vietnam.” (Currey 2005, 275).

93 This decision was undertaken in the belief that NVA formations could force the removal of the Marine
base at Khe Sahn. (Duiker 1995, 210).

89

of reunification (Yearbook on International Communist Affairs, 373). More tangible
efforts at curbing dissent were also employed as Le Duan helped orchestrate the purge of
domestic political actors less favorably disposed to his war policies.94 The summer of
1967 witnessed further dismissals of leading communist officials (Ang 2002, 119-120;
Durkier 1996, 65-66).95 Silencing dissenters Le Duan galvanized support for his
preferred war policies. Dismissing earlier failures of large unit engagements with
American forces, the executive was able to mobilize for a high-intensity campaign in the
RVN.

The pro-Beijing/large unit faction and the executive still embraced a Maoist
archetype of a large unit strategy as a vehicle to rapidly achieve final victory. An
attritional strategy was endorsed that would attempt to hold RVN urban centers long
enough to catalyze a popular revolt amongst urban workers. After having infiltrated the
cities with small commando detachments, larger VC and NVA formations would follow
up to secure strategic positions gained by the uprising.

Having ensured the political collapse of the Saigon government and the defection
of most of the South’s army they could demand substantial issue-goods in return for the
extrication of US military forces stationed in a hostile country (Post 1994, 88). The plan
for a General Offensive was certified in April 1967 and is better known as the Tet

Offensive (Abuza 2001, 61-62).

 

9” “A primary factor in the lack of change in the Politburo leadership’s views between the 1965 and 1968
crises was, of course, that the Tet offensive constituted the first major crisis of the war directly initiated by
Hanoi. The Tet offensive constituted a crisis for Hanoi because of its aftermath: it did not achieve decisive
military goals.” (Rogers 1976, 121).

95 “After Thanh’s death [a pro-large force NVA commander in the south] in July 1967 Giap replaced him.
Despite hi disagreement with the Tet Offensive concept he was placed in charge of its execution.”

(Ang 2002, 119).

90

The objectives of the general offensive were in no way reached and by nearly all
accounts the communist forces were crushed. In spite enjoying the element of surprise
and having successfully baited a number of American combat battalions into protecting
the besieged Marine outpost at Khe Sahn along with other border areas, the results were
devastating. In the first three weeks alone some 45,000 communist soldiers were killed
(Clodfelter 1995, 135). More importantly many cadres and communist activists had
emerged in order to orchestrate the rebellion only to be destroyed en masse in the ensuing
combat (Currey 2005, 304-305).

The Tet Offensive attacks as well as the follow on attacks launched in May 1968
produced a major backlash against the executive’s war policies. The General Offensive
not only failed to catalyze a revolution, but incurred massive costs in personnel. This
was particularly significant given ebbing willingness of North Vietnamese populace to
mobilize.96 The probability of coercing the US out of the South through destruction of
the Saigon regime now seemed much more remote and the span of time until
reunification could be achieved appeared longer than ever.

The results of the battlefield had clearly revealed the disposition of capabilities
between the US military and the NVA. The failed General Offensive had been
supremely costly, yet the follow on attacks were launched in May even after these losses.
The executive’s preferred strategy was now assessed as being massively inefficient

compared with a low intensity war (Yearbook on International Communist Affairs 1970,

 

’6 Several articles in official publications from 1968 that illustrate the debate amongst domestic political
actors in terms of the need to build up the North as a safe base area for continuing the effort for
reunification in the south. (Yearbook on International Communist Affairs: 1969, 883-884).

91

11-14).97 A reduction in the political fortunes of the executive was met by changes in
war policies. Faced with the potential for deselection and the reality of diminished
political authority, Le Duan was compelled to move away from his preferred military
strategy.

Mounting political pressure on Le Duan and the prospect of further reduced
political clout or deselection led to a greater willingness to engage in negotiations. By
heightening the likelihood of achieving the issue-goods at stake in the war, an American
military withdraw, might substantially alter the utility of domestic political actors.
Hastening the realization of benefits through a quicker settlement would also reduce the
discount being applied to the realization of a reduced level of issue-goods benefits.

The subject of negotiations with the US was not a new idea, however very little
accommodation had been made to American efforts to diplomatically resolve the war.
Generally the pro-Moscow/North-first cohort took a more favorable stance with respect
to negotiations.98 The protracted war oriented domestic political actors now led by
Truong Chinh stressed the importance of draining the Americans of their willingness to

fight rather than undertaking the costs of direct military confrontation (Post 1994, 105).

 

97 “Inevitably the disappointment of Tet rekindled the strategy debate. In May of 1968 issue of the Journal

Tuyan Huan, Le Duan once again tried to justify his ‘offensive’ strategy. By July, however, other
newspaper articles showed that he was on the defensive. . .Later, in a National Day Speech, Premier Phan
Van Dong made brief meaningful reference to protracted war. And within the highest circles of the North
Vietnamese government, another critical voice—that of Truong Chinh—was gaining new authority.
During the summer of 1968, following the Tet Offensive, Truong Chinh’s star was definitely on the rise. It
was not a meteoric ascent—Le Duan had not been completely repudiated and the debate over strategy
continued. The dispute was serious enough to delay the publication of Truong Chinh’s historic speech on
protracted war commemorating the 150'h anniversary of Karl Marx’s birth. . .From Truong Chinh’s speech
it became clear that Le Duan faced a new and formidable rival—and that Giap’s thesis had prevailed. The
address echoed in almost classic terms the arguments and prescriptions. for protracted war.” (Yearbook on
International Communist Affairs 1970, 11-14).

9’ With detente underway it suited the Kremlin’s interests to prod Hanoi toward a settlement (Kimball
2004, 14-15).

92

Figure 5: Distribution of Elite Opinion in North Vietnam

Beginning of American War After Tet Offensive

Executive’s policy . . .
preferences/ Distribution of Distribution of
cm'rent war Domestic Political

, , Domestic Political ,
polic1es Actors preferences

Actors’ preferences Executive’s policy

preferences/ ‘5
current war policies

 

 

 

 

 

 

< > < #
L855 favorable More favorable Less favorable More favorable
relative utility relative Utility relative utility relative utility

By contrast the South-first/large unit faction viewed negotiations more darkly, as
Beijing advocated the broaching of negotiations only from a position of military strength
(Zhai 2004, 283). Le Duan would only seriously entertain negotiations when
preconditions were met, including a cessation of the Rolling Thunder bombing campaign
and a large American troop withdraw (Ang 2002, 114).99

The dynamic changed somewhat when Johnson’s own political implosion resulted
in a bombing halt over most of North Vietnam. Even with these concessions to Hanoi it
was only after the application of further political pressure and lobbying that negotiations

were agreed to by North Vietnam’s executive (Thai 1985, 64).

 

’9 By one account the American officials and proxies had made at least 2,000 individual efforts at
negotiating with Hanoi between 1965 and 1968. (Goodman 1978, 24).

93

However the differences between domestic political actors and the executive was
over means not ends and negotiations were seen by most domestic political actors as a
means of eroding American resolve rather than a forum for compromise (Ang 2002, 138-
139).loo The principal sticking points in the negotiations in Paris was withdraw of NVA
soldiers from the South, which was not an option that Hanoi took seriously (Prados 2004,
3 5 8-36 1).

The movement away from the aggressive attritional strategy exemplified during
the General Offensive did not preclude major combat from occurring. In 1969 the NVA
and VC combined for over 130,000 combat deaths (Clodfelter 1995). While the United
States was also encountering a reduced willingness by the public to mobilize, a new
American military strategy under General Creighton Abrams was becoming more
effective (Sorley 1999). These pacification efforts expanded Saigon’s control over the
RVN’s populace, facilitated in large part by the VC’s degraded political infrastructure
destroyed during Tet.

With negotiations stalled and costs of the war rising, demands for changes in war
policy continued amongst Hanoi’s elites. Facing another bout of recurring political
pressure undermining the executive’s position, a retrenchment of the large-unit strategy

was decided upon in February 1970 (Rogers 1976, 55-56, 121-122).'01 The executive

 

"’0 The Paris Peace Talks between the US and North Vietnam thought that the Communist delegation used
negotiations instrumentally in order to demonstrate a willingness to compromise but did not seriously
entertain substantially altering their bargaining position for a compromise settlement (Ang 2002, 138-139).

101 “ . . .Le Duan announced a policy shift in a major speech at the Party’s fortieth anniversary. He
advocated more attention to economic needs to the north, needs ameliorated by the cessation of US air and
naval attacks north of the twentieth parallel late 1969 through 1971, the new Party line was to give equal
priority to restoration of the base of socialism in the North and the war effort in the South. As later became
evident, this change did not represent victory for the conservative Truong Chinh viewpoint but was a
necessary tactical pause to reconstitute the resources needed to continue the policy of mounting another
violent surge, the PAVN [People’s Army of Vietnam] offensive in the South in 1972.” Rogers 1976, 121-
122

94

scaled back the war effort in the South in order to accommodate the North’s economic
development. The policy shift addressed concerns of domestic political actors regarding
the viability of the regime at of the expense of increasing the time until the realization of
benefits.

As the communist movement in the South enervated the mobilization of local
resources grew increasingly difficult, placing more of the burden upon the North (Currey
1997, 304-305). This served as the chief cost to the protracted war/North-first domestic
political actors, lobbying efforts continued most prominently by Giap’s article detailing
the benefits of a guerrilla warfare (Currey 2005, 305). Still the DRV’s issue-goods
negotiation position did not substantially change, political pressure was applied to the
executive to change how the war was fought not the issue-goods it was being fought for.

Understanding that Hanoi was simply biding its time until the US pulled out of
the South and frustrated by the lack of progress in negotiations Nixon ordered a US-
ARVN attack on communist sanctuaries in Cambodia. Starting in April 1970 the US-led
incursion destroyed substantial amounts of material but made little contact with NVA
forces there. The American executive decided to launch a second offensive in February
1971 designed to disrupt communist logistics in Laos (Clodfelter 1995, 176-179). Unlike
the Cambodian invasion, the Laotian operation was executed almost exclusively by the
ARVN. After initial progress, a NVA counterattack routed the South’s units leading to
an embarrassing setback for Vietnarnization as US forces continued to withdraw.

The success of the NVA in combat with the ARVN indicated the increasing
likelihood of the defeat of Saigon once the Americans left. Despite the success at turning

back the ARVN operation in Laos, the costs to the stability of the regime remained steep

95

to the North-first/protracted war faction. '02

The broader society was increasingly weary
necessitating the state’s use of more coercive methods to stop draft dodging and other
failures to mobilize (Yearbook on International Communist Affairs 1972, 682).

The declining resolve of the populace was interpreted differently by the South-
first/large unit faction and the executive, 1972 was a US election year something that had
not escaped the Hanoi’s attention. Nor had Nixon’s efforts for détente with the Soviets
and thawing relations with Beijing (Randolph 2007, 25). The US was clearly trying to
use its new-found diplomatic clout to help pressure the recalcitrant Hanoi toward a
negotiated settlement. Le Duan decided to order another major offensive.

The North-first/protracted war cohort of domestic political actors was dismayed at
the suggestion of yet another general assault on the South. Preferring instead to wait out
the US and then strike. The executive by contrast continued to heavily discount the

utility of benefits by the anticipated time until they could be realized.103 Inhibitions of

domestic political actors to endorse the plan invited another effort at elite repression in

 

102 “The year 1972 for North Vietnam was one of extraordinary strain as new pressures developed or old

ones worsened in virtually every sector of society. . .Leadership system pressure, developing from doctrinal
differences of long standing and now involving a particularly stubborn divisiveness deep into the ranks of
Party members in the North (the Lao Dong) and the South (NLF). Internal social or psychological
pressure. . .which is seen by the regime as a near permanent condition of insufficient élan and ominously,
as the rise of counter-revolution in North Vietnam.” (Yearbook on International Communist Affairs 1973,
1-2).

"’3 “. . .the Politburo was heartened by ARVN’s debacle at Lam Son 719 [ARVN invasion of Laos]. Early
summer, not long after Giaps’s victory in Laos, making members of the northern government authorized
him to launch a major assault against the South sometime the following year, while the United States was
still drawing down its forces, rather than waiting until after the departure of American units. ARVN would
be no threat, and an attack would humble not only the South but the United States as well. They could
smell victory, and in any case, war should end with a bang, not a whimper.

Giap listened to all this with mounting dismay. He had believed such desires were a thing of the past, dead
and buried. Hoping to find a way to short-circuit such plans, he enlisted the aid of Truong Chinh, and both
men vigorously opposed mounting a southern campaign. As at Tet in 1968, Giap believed it was the wrong
time to undertake such a major offensive against the South. It was still too strong. ARVN had improved
too much as a result of Vietnarnization, American field forces might be fewer in number, but US air power
could still mount savage strikes against ground forces. An offensive would drain the North of too many of
its units and its meager supplies. One again, Giaps’s position did not convince others in the Politburo, and
he was ordered to prepare a strike against the South and to achieve a battlefield victory that would enable
Ha N01 to dictate the terms ending the war.” (Currey 2005, 283-284).

96

order to prosecute Le Duan’s preferred policies (Yearbook on International Communist
Affairs 1973, 561).’°" The spring offensive went forward.

The invasion plan called for an offensive that was to be much more conventional
than that of the 1968 attacks. Employing mechanized forces, vast quantities of Soviet
arms were shipped to the Democratic Republic including heavy artillery and armor. The
assault would be main force army units utilizing another campaign to seize RVN territory
and foment the political collapse of Saigon (Randolph 2007, 27).

The Easter Offensive, as the campaign became known, was initially successful.
ARVN installations along the DMZ were pounded by Soviet-made artillery and quickly
fell. The northern provinces of the RVN were be conqured and many ARVN units
panicked and fell back southward. Additional NVA forays towards Saigon originating
out of Cambodia were launched in order to busy Saigon’s strategic reserves. The
American effort to protect South Vietnam looked to be in serious jeopardy and with it the
US personnel still stationed in the RVN. Not to be dissuaded by warming relations with
Beijing and Moscow, Nixon had no intentions of losing the war on his watch.

American advisors helped to organize a defense and more competent ARVN
officers were assigned to the front. However, the outcome of the Easter Offensive was in
large part decided by the massive reintroduction of attack aircraft the United States
directed to aid Saigon. Nixon also unleashed air attacks upon previously restricted Hanoi
and Haipong along with mining the coast (Randolf 2007). The crushing weight of

American airpower crippled the supply infrastructure of the now more logistically

 

1°“ “There are indications that the VWP [Vietnamese Workers’ Party, Lao Dong] conducted a minor purge
during the middle of 1972. In addition to a number of articles in the Hanoi press concerning the treatment
of dissidents and counter-revolutionaries, there was a Politburo directive on ‘developing criticism of the
press’ Hoc Tap, the VWP monthly theoretical journal, explained on its’ April issue.” Yearbook on
International Communist Affairs 1973, 561.

97

consumptive NVA formations. The more easily spotted communist units fighting pitched
battles were also devastated by tactical strikes, the initially promising Spring Offensive
ground to a halt.

Again domestic political actors witnessed heavy casualties that defeated their
intended aims. A necessary retrenchment of main force attrition offensives also
compelled a reexamination of the North’s negotiating strategy. In the aftermath of the
failed Easter Offensive and the enormous costs inflicted upon the NVA, Giap is said to
have advocated modest changes in negotiating strategy in order to hasten the American
exit from the conflict (Weil 1975, 464). Le Duan would again be compelled to satisfy
these demands in order to secure his position in the Hanoi Politburo.

At times he was forced to apply a less aggressive military strategy in order to
accommodate the demands of the rival faction in Hanoi. Even though war policies would
revert back to his preference, accommodation to political pressure was made. Whereas
political pressure went unaddressed in Japan and for much of the Iran-Iraq War, these
executives were deselected or lost much of their political authority. As with Mao’s
handling of the Korean War, Le Duan’s accommodation of demands for alternative
policies allowed him to avoid sanction by his elite constituents.

The battlefield informed both the executive and domestic political actors as to the
relative resolve and capabilities of the combatants. However, the interpretation of costs
and benefits of the war’s utility was filtered through the preferences of the observers.
The DRV’s executive enjoyed strong support amongst a plurality of domestic political
actors for his preferred war policies. Another faction adamantly opposed his military

strategy and lobbied for the use of negotiations as a means of removing Hanoi’s lethal

98

opponent from the war. The level of debate in the confines of the Lao Dong Party
recommends itself to an alternative interpretation than a unitary assessment as to the
prospects of victory given battlefield outcomes.

An alternative explanation would be to suggest that war policies were derived
from the median position of Hanoi’s elites. This however lends too much credence to the
“collective” nature of the DRV. Instead it appears that Le Duan drove policy and was
dissuaded from his preferred course for unification with the South not by the council of
domestic political actors, but by the political pressure they applied to him.

The critiques of the war were likewise directed at the way the war was being
conducted defined by the ideological cleavage within the Politburo. Changes in military
strategy were registered but these were not necessarily driven by an assessment of key
variables. The aggressive attritional strategies of the executive were not stopped by
acceptance for the need to change tactics because of high losses, but by political pressure
and the complete exhaustion of Hanoi’s forces.

The Tet Offensive and the Easter Offensive also did not constitute an effort by Le
Duan to gamble for resurrection in that the aims of the war never changed. The garnbles
indicated a riskier strategy, but were not made out of a fear of future punishment. Rather
it was an attempt to more rapidly engender the inevitable collapse of the Saigon regime.
Unlike Japan, China, and Iran the question over issue-goods position is never really the
focal point of policy debate. Instead of the issue-goods being pursued being the crucial
point of contention, it is the level of resources that are to be devoted in what is likely to

be an extended struggle that the debate between principals and agent contend. The

99

danger to the promotion of domestic political goals in North Vietnam becomes a source

of political pressure.

Germany: World War [I

The Third Reich’s military successes are almost unprecedented in the modern era.
The speed and scope of its victories were only equaled by the enormous balancing
coalition these conquests invited. Hitler had redeemed Germany’s Teutonic pride
through the evisceration of the Versailles Treaty, but his megalomania began to cost the
nation dearly as inexorably enemies closed in.

Germany’s domestic political actors consisted of individuals that held top
positions in the Nazi Party or within the government. Other members included high-
ranking military officers in both the Wehrmacht and Waffen SS, leaders of the SS
(Protective Squadron) along with its sub-department the Gestapo.

The ability to repress the public was more than adequate and the Nazi state had
been detaining political prisoners for years and instilled fear in most of the public.’05 By
contrast the repression of elites was not nearly as common. Unlike Stalin’s Soviet Union
where general officers and leading party members were purged and executed with some
frequency, Hitler employed different means of maintaining power.

Many domestic political actors were doctrinaire National Socialists and largely
shared Hitler’s perceptions of World War II’s costs and benefits, or simply thought they

would share his fate. These elites did not apply much in the way of political pressure to

 

"’5 The propaganda of Goebbels was also considered to enjoy a high degree of efficacy. Additionally Hitler
did enjoy some genuine popularity, particularly throughout the early part of the war (Herzstein 1979).

100

alter war policies until very late in the war. Other elites, mainly leading military officers,
read the signs of disaster and attempted to alter Hitler’s war policies.

Some military leaders were co-opted into the endorsing Hitler’s rule as much as
they were sanctioned (Goda 2000).'°° Side payments to Wehrmacht officers who
performed well in the executive’s eyes took the form of cash and promotions. Officers
that refused to obey orders or seemed less than exuberant about Hitler’s plans were fired
or compelled to resign (Irving 1977, 66-67). For the most part these sanctions were
usually meted out as a result of some specific policy dispute or clash of personalities
rather than out of a perceived threat to the executive’s position.

While rearmament under the Nazi regime had increased the resources and prestige
of the Wehrmacht, Hitler was never esteemed by many senior officers (Hoffman 1988,
71-86; DeWeerd 1949). State ideology had been infused into the armed forces, but a
large portion of the seemingly sycophantic officer corps harbored disdain for the
executive. Hitler’s lack of military pedigree led some officers to deride him privately by
his World War I rank of “corporal”. More tangibly the Prussian military aristocracy
became increasingly concerned about his micromanagement and grandiose ambitions
(Magenheimer 1999, 121).

As early as the Nazi absorption of Czechoslovakia in 1938 and before the
operation to conquer France in 1940, some officers contemplated removing or killing
Hitler. These were usually lonely attempts that did not gather the requisite critical mass

of domestic political actors to make serious coup or assassination attempts possible

 

'06 Side payments to domestic political actors can act as a key factor in assuring their loyalty (Goda 2000).
If domestic political actors receive side payments fi'om the cin'rent executive they should be less inclined to
deselect and risk the payoffs coming flour a new executive (Bueno de Mesquita et al. 2003).

101

(Deweerd 1949, 201). '°7 Comparatively few domestic political actors were willing to
take the personal risk of usurping the executive given the territory garnered through
Hitler’s aggression and the yet largely unrevealed monstrosity of his ideology.

The executive’s insistence on a Blitzkrieg strategy during the assault on France
illustrated his cunning and vision (Mearsheimer 1999).“ Moreover the easy conquests of
most of Europe enhanced his prestige as well as afforded Germany a massive net utility
gain in terms of economic resources, political clout, and other issue-goods (Mearsheimer
2001, 20). Consequently, early attempts against Hitler received little support amongst
domestic political actors in that the net utility of the war was comparatively high. ”9

Victories in 1941 continued to vindicate Hitler’s war policies, but the executive’s
drive for expansion floundered in the vastness of Russia. The defeat of Operation
Typhoon directed at Moscow in the fall-winter of 1941 exacted heavy casualties. But in
the winter of 1941-1942 the German executive again seemed to have willed his nation to
victory. Proclaiming stand-fast orders to his officers, the executive had helped hold the
line in the Soviet Union after the Red Army’s brutal counterattack (Irving 1977, 355-
366)

Surviving into the spring with its army largely intact, Germany was still faced
with an open ended war on the Eastern Front. In order to break this deadlock in the
summer of 1942 a huge German offensive was launched into the Caucuses. Operation

Blau was intent on depriving the Soviet Union of its oil supply and potentially securing a

 

107 Amongst the leading echelons of German officers it was important to have a fairly large contingent of
military men in agreement in order to credibly supplant Hitler. Moreover, this had to be done in such a
way as to garner the support of the broader public (DeWeerd I949, 201).

'0’ Starting in the December 1941 Hitler took over direct control over the military establishment. (Irving
1977, 360).

"’9 Halder, the chief of the German General Staff of the Army, in 1940 was contacted regarding
preparations for a coup. He responded that another serious military setback would be necessary in order to
secure broader support. (Hoffman 1988, 93).

102

minority uprising against Moscow (Magenheimer 1999, 147-153). The German army
that suffered greatly in the previous winter seemed to have regained its potency driving to
the Volga River.

Most of Stalingrad was captured in a series of infantry assaults, but the long
campaign into the steppe had overextended the German’s position resulting in
comparatively weak Axis units from Romania and Italy protecting the flanks. Late in the
fall huge armored formations of the Red Army smashed the poorly-equipped Axis units
north and south of Stalingrad. The Soviet counterattack led to the encirclement of the
German 6th Army inside the city.

In the face of the greatest crisis in the war to that point, the German executive
refused to relinquish the costly conquest of Stalingrad. Overruling the German high
command, Hitler ordered that the encircled army fight in place rather than trying to
breakout and regain their own lines. A Wehrmacht counterattack that designed to break
the encirclement and rescue the 6th Army failed. The efforts to supply 250,000 men by
air proved to be completely inadequate, resulting in the entire force being captured
(Erickson 1975a, 472).

Upon defeating the Wehrmacht in Stalingrad the Soviets initiated a number of
offensives against their staggered opponent, driving the Germans from the Caucuses.
Massive defeats in the Soviet Union were accompanied by the loss of North Africa,
which foreboded an Allied invasion in southern Europe sometime in 1943 (Magenheimer
1999,176)

The Wehrmacht’s leadership understood what the vengeful Soviets and the

Anglo-American declaration of unconditional surrender would mean for their nation.

103

These domestic political actors knew the loss of irreplaceable men and material meant a
reduced likelihood of preserving German sovereignty. Political pressure emerged as
senior officer began to see the micromanagement of strategy by the executive as costing
the regime dearly in terms of military assets (Irving 1977, 415-416). The comparatively
negative cost-benefit calculations of many domestic political actors resulted in a growing
conspiracy to deselect Hitler (Hoffinan 1988, 109).

Whether from paranoia or intuition the German executive had a sense of the
growing threat of deselection especially amongst high-ranking officers.”° Understanding
that it was unlikely that Germany could destroy the Soviet Union Hitler looked to strike a
blow that could compel some intermediary settlement with Moscow (Healy 1992, 7).

The executive contemplated a resurrection of blitzkrieg through an encirclement
operation of major Red Army formations surrounding the city Kursk.

While the German high command appeared to agree that the strategic initiative
could not wholly pass to the Soviets, it was Hitler who again demanded on striking first
(Newton 2003, 7-11). Rejecting options to let the Red Army strike first and then launch a
counterattack to gain the initiative, the executive demanded that relinquishing territory
for a mobile defense was unacceptable (Dunn 1997, xv).

Hitler thought a victory at Kursk might allow for enough of a respite to be able to
deflect an Anglo-American invasion in Western Europe (Magenheirn 1999, 203-204). If
the Soviets could be checked and military assets redirected against the Western allies,

perhaps the enhanced prospects for a more limited set of issue-goods would mitigate

 

”° Hitler had an abiding distrust of the senior officer corps and long feared that they were conspiring
against him. (Irving I977, 50, 660; Quinnett 1978).

104

burgeoning political pressure. At the very least it would permit Hitler the opportunity to
garner greater issue-goods benefits through a treaty recognizing German territorial gains.

The assault on Kursk, July 1943, ended in a stalemate and destruction of some of
Germany’s most potent panzer formations. The failed campaign also demonstrated that
even the more limited goal of inflicting a debilitating blow to the Red Army, as opposed
to conquering the Soviet Union, was now impossible. Afterward the Red Army never
relinquished the strategic initiative and continued their drive to eject the Wehrmacht from
the Soviet Union (Erickson 1975b).

While the Third Reich still controlled most of its earlier conquests, in the second
half of 1943 the war’s momentum had decidedly turned against Germany. The likely
issue-goods that Berlin was to receive should the war continue was becoming
substantially less in the eyes of many domestic political actors. Even Joachim von
Ribbentrop, the foreign minister and ardent Nazi, went against party doctrine by
suggesting that some negotiated settlement be offered to the Soviets in the wake of the
failure to strike a devastating blow in the assault on Kursk (Irving 1977, 567-568).

Other elites joined Ribbentrop such as Field Marshall Erich von Manstein, to
lobby for a more accommodating approach to the Russian people and by extension the
Soviet government (Magenheirn 1999, 193-195). Hitler did not act on this advisement
considering the territory already conquered too valuable and negotiations a sign of
weakness. These efforts at altering executive policy went nowhere, offering no
accommodation to the building political pressure.

Sensing a challenge to his authority Hitler attempted to further indoctrinate-

repress the Werhmacht with a new breed of political officers. In December 1943, Hitler

105

ordered the creation of political officers to be integrated into the German armed forces
(Quinnett 1978). Intensifying efforts to limit the niunber of military elites that could
attempt to deselect Hitler from power. Only in January 1944 did the German executive
acquiesce to a negotiated settlement, offering peace with the Soviets in exchange for an
independent Ukraine and raw materials for the ongoing war with the Western Allies
(Erickson 1975b, 162). This was not an offer that Stalin was going to seriously consider.

Hitler’s reticence to appreciably alter Germany’s bargaining position to
something amenable to the Soviet Union served as a major source of discontent with
current executive war policies. The perceptions of the war’s utility were even further
degraded after the invasion of France. The Normandy landings in June 1944 signaled a
new shock to the cost-benefits analysis for many domestic political actors.

The well-planned D-Day invasion left little to chance and the strength of the
Western Allies was quickly apparent. Having breached the most important physical
boundary protecting German conquests from their would-be liberators, the probability of
a successful continental defense rapidly depreciated. In July 1944 the 7th German army
was facing encirclement, in response another series of stand fast orders had been made by
Hitler rather than permit retreat (DeWeerd 1949, 207). As with similar orders on the
Eastern Front these actions ensured the destruction of yet more irreplaceable formations.

Military officers now realized that to salvage anything short of the unconditional
surrender promised by the Allies, they must act quickly. It seemed to some domestic
political actors that a possible accommodation might be to exchange territory still
controlled by Germany for a peace settlement ensuring the state’s sovereignty (Yelton

2000, 1070). However, it was widely believed that Hitler could not be counted upon to

106

express sympathy for this agreement. Perhaps more importantly, the Western Allies had
indicated that they would not negotiate with him in office regardless (Hoffman 1988, 95-
105). Facing a situation where current war policies were drastically reducing perceived
utility and with an executive who refused to bend to political pressure, a collection of

domestic political actors struck.

Figure 6: Distribution of Elite Opinion in Germany

- After D-Day-
Before German Invasron of
France Before Army Purges
Executive’s policy Distribution of Distribution 9f
Domestic Political

preferences/
current war policies

\

 

Domestic Political
Actors’ preferences

/

 

‘_

 

5
Less favorable More favorable
relative utility relative utility

Actors’ preferences

 

 

 

Executive’s policy
preferences/ ———>
current war policies
a +7
Less favorable More favorable
relative utility relative utility

While earlier assassination attempts had been made in 1943 and 1944 these efforts

were comparatively simple compared with the 20 July 1944 plot (Hoffman 1988).

Colonel Claus von Stauffenberg, despite his comparatively junior rank, was a central

figure in the conspiracy against Hitler and planted the bomb that failed to kill him. The

oft suspected political pressure was violently evidenced.

The conspiracy’s members, overwhelmingly represented by high-ranking military
officers, emerged to launch a coup against the Nationalist Socialist government. The
failure to assassinate Hitler meant that they were not facing a confused opposition in the
remaining Nazi hierarchy. Instead the effort at seizing control in Berlin failed and was
followed by a gruesome targeting of the coup’s participants and those suspected of being
fellow travelers.1 1’

The executive understood that he could not destroy the Wehrmacht officer corps
without a serious reduction in military capability, nor could Hitler be completely sure that
repression and oaths of loyalty would hold (Knox 2000). A decent peace settlement must
be achieved before the war was completely lost or domestic political actors succeeded in
removing him. In order to address these challenge both repression and an alteration in
strategy were used.

Hitler began to enforce a further Nazification of the military (Yelton 2000, 1072-
1073). This included both reshuffling personnel in charge of different portions of the
armed forces as well as further nationalist socialist indoctrination of soldiers. The push
for greater loyalty was an extension of a broader program of mobilizing the population as

a whole.112

 

l" An estimated 700 German officers were executed as result of the failed coup (Deweerd 1949, 207).

”2 “Among the welter of psychological mobilization programs launched or expanded in mid- to late 1944
was the effort to ferret out defeatists in the officer corps and replace them with dedicated National
Socialists. The naming of Himmler [head of the SS] as the new commander of the Replacement Army was
only the most public manifestation of this trend. Hitler expected the Reichsfuhrer not only to restore the
loyalty of the force which had provided the bulk of active participants in the 20 July plot, but also to
supervise the intensified political indoctrination of new recruits. In addition, a major expansion of the
National Socialist Leadership Officer program aimed at fanaticizing the existing Wehrmacht, thereby
improving morale and battlefield performance which inoculating the military against further anti-Hitler
conspiracies. To infirse further zeal into local defense and to make certain that the Wehrmacht fought
where it stood. . .there was the creation of the German Volksstrum, a centerpiece of the new strategy both
because of its size and its dual role as a tool for mobilization and motivation of German civilians.” (Yelton
2000, 1072-1073).

108

The final measure of German resources was extracted for the war effort, including
the conscription of previously ineligible soldiers and the closure of non-essential
industries (Delaforce 2004, 14). Despite strategic bombing and other disruptions
industrial production in Germany actually peaked in 1944 (Magenhiemer 1999, 234-235).

These new resources would be used create a series of strong-points called the
Western Wall of the Reich increasing Allied casualties by forcing them to attack prepared
positions (Yelton 2000, 1069-1070). The intention of this fierce defense of territory with
isolated groups was to deny the comparative advantages of speed and mobility to the
Anglo-American forces. A successful Allied maneuver strategy might afford
comparatively less costly and achieve demoralization of the Wehrmacht (Yelton 2000,
1070). After exacting an expensive toll in casualties upon the Allies, Hitler intended to
strike back with the last of his mechanized reserve.“3

The objective of the Ardennes offensive was to attack the political will of the
Anglo-American, separating and encircling them in the hopes that increased causalities
might fracture what he considered a fragile partnership (Irving 1977, 741).”‘ Hitler
believed that a massive encirclement of Allied forces would compel Churchill to
negotiate or even force him and/or his American counterpart out of office (Delaforce
2004, 29). An assault using a maneuver strategy could enhance the likely issue-goods

that could be obtained (Hoyt 1988, 327).115

 

"3 While definitely circumstantial, all secondary narratives of the Battle of the Bulge (Ardennes) examined
for this case study begin by recounting the events of 20 July 1944.

1” Hitler believed that such politically disparate a coalition as the Soviet Union, United States, and Great
Britain was inherently fragile and that they could be split apart and dealt with separately (Strawson 1972,
8).

"5 “For months Hitler had been talking about striking the western Allies a blow that would turn the war on
that front completely around. After the failed assassination attempt of July 1944, his attention to this goal
became an obsession. Something had to be done to relieve the two-front pressure.” (Hoyt 1988, 327).

109

Understanding the casualty sensitivity of democracies, it was hoped that
concessions could be extracted from the Allies in order to avoid massive costs in the
process of invading Germany itself (Yelton 2000, 1070).”‘5 A political settlement would
then improve the utility calculations of those dissenters not yet rooted out by salvaging
peace terms leaving Germany in a comparatively favorable position. Even if the Anglo-
American coalition did not fall apart or immediately negotiate after another Dunkirk, then
it might still buy Germany time to ramp up its production of advanced weapons like the
V-2 ballistic missile and Me-262 jet fighter (Yelton 2000, 1069-1070; Nobecourt 1967,
69-72). Hitler thought that these “vengeance” weapons once in wide use could
potentially turn the tide of the war.

Extreme precautions for secrecy had been undertaken such that reports of German
panzers crashing through the lightly guarded French frontier were a source incredulous
shock to the Allies. Aided by poor flying weather the Germans raced to sever
communications and drove toward Antwerp in order to recapture one of Allies’ few deep
water ports on the continent.

Strained by a shortage of fuel and facing fresh Allied reserves the German assault
stalled. A break in the winter weather then permitted the potent Anglo-American tactical
aircraft ready targets as the exposed panzers struggled forward. Despite early successes

the offensive was doomed to failure and the last of Germany’s armor reserves were spent.

 

”6 “As Himmler [Head of the SS] indicated, the purpose of all this bitter, often militarily senseless

resistance was to maximize enemy, particularly British and American, casualties in order to erode and
ultimately collapse Allied morale. Once the enemy’s will to resist had broken, Hitler expected the Allies to
sue for peace on Germany’s terms. He had no intention of requesting negotiations; the example of
Versailles had taught him the danger of seeking peace terms from a position of weakness. Nazi ideologues
considered the Allied coalition of capitalists and communists to be inherently unstable, and they hoped to
hasten its inevitable fragmentation by this prolonged resistance and casualties which it would entail.”

(Y elton 2000, 1070).

110

The initiative lapsed to the Allies for the remainder of the war as they mopped up the
floundering Wehrmacht.

Hitler remained as executive for remainder of the war as elite repression had
succeeded in cowing dissent. Moreover, prominent leaders in the party and elsewhere
regarded the stated Allied terms of unconditional surrender as binding and after the leak
of the Morgenthau Plan there was apparently little to be gained in attempting to deselect

the executive.“7

With no perception of improving the issue-goods position of the Reich
these leaders did not hazard a serious attempt to deselect the executive.

Hitler did alter strategy in an attempt to improve the utility of further fighting.
Military strategy does not appear to be changed in accordance with a rationale assessment
of key variables, such as casualties or enemy losses. Instead the strategy toward the end
of the war is better described as capricious. The assault on Kursk and the Ardennes
Offensive were both designed to dramatically enhance Germany’s prospects in the war
and improve the perceived benefits of domestic political actors. To a large extent war
policy in Germany reflected the preferences of the executive, but also evidenced changes
in strategy that reflected the need for assuaging elite concerns over war policy.

Some evidence of the risk acceptant strategy is also found. While Nazi Germany
was not a semi-autocratic regime, it appeared that Hitler believed that he had take
military garnbles in order to change the course of the war. At both Kursk and in the

Ardennes aggressive operations were attempted which failed. While the Reich was

heavily mobilized for much of the war, more drastic measures were taken toward the end

 

”7 The Morgenthau Plan was a blueprint for post-war Germany that included the de-industrialization of the
state and other harsh terms, likely more destructive than had been the Treaty of Versailles (Irving 1977 ,
739-740).

111

of the conflict in order to increase the costs on their opponents in hopes of securing a
stronger issue-goods negotiation position.

For most of the war Hitler was able to execute his preferred war policies. Due to
a strong ideological affinity for the executive political pressure was muted amongst a
large portion of domestic political actors. However as a substantial cohort domestic
political actors began to view the war’s utility in starkly negative terms intense political
pressure emerged, particularly from the German officer corps as the prospects for the
regime’s survival dissipated. Eventually the executive hazarded deselection several times
as a result. This might tentative evidence for the unitary actor assumption, but this would
discount the importance of political pressure in the decision making process as well as
miss the threat of deselection that hung over the executive throughout the war.

Hitler made occasional peace offers, to both the Soviet Union and later the
Western Allies. However these issue-goods positions were not realistically going to be
accepted by Germany’s antagonists. Consequently the prospect of deselection increased
as a growing number of domestic political actors demanded alternative war policies. The
refusal to accommodate political pressure nearly cost the executive his life on several

occasions.

Iraq: Iran-Iraq War
Iraq was the initiating state in one of the most bloody interstate wars of the
Twentieth Century. While predicated on a readjustment of an earlier territorial

agreement, likely this masked a more pressing imperative to cripple the Islamic

112

government in Tehran. Instead of a quick victory the war would become a highly
personalized struggle for Hussein’s political survival.

The high-ranking military officers, leaders of the state’s repressive apparatus were
influential leaders in Iraqi politics. Additionally elite members of the Baath Party and its
affiliated institutions such as Revolutionary Command Council (RCC), the cabinet, and
the Regional Baath Party Council constituted Iraq’s domestic political actors (Tripp 1993,
97; O’Ballance 1988, 91).“8

Despite Hussein’s honed capacity for repression of Iraqi society, there was a
serious fear that the Shiite revolution in Iran could spread to his state (Chubin and Tripp
1988, 54). Settling upon a quick territorial strike that would demonstrate the
vulnerability of Khomeini’s regime to internal opponents he launched the 22 September
1980 assault (Karsh 1987-88, 89). The invasion resulted in territorial acquisition but at a
greater cost and longer duration than was originally expected.

While armed with fairly sophisticated weapons, the Iraqi army was not able to
employ them to their full potential. Also a demonstrable lack of skill by senior level
officers denied the Iraqis the benefits of enjoying the strategic initiative (Herzog 1989,
259-260). Aside from these military failures, the lack of enthusiasm amongst Arabs and
other minorities living in Iran for the Iraqi invasion meant that solidifying territorial gains
would be more difficult (O’Ballance 1988, 36-37).

What was designed to be a maneuver style strategy stalled as combined arms
tactics were not be effectively used and key logistical hubs that could have cut off Iranian

militias from reinforcement were not seized (O’Ballance 1988, 32-33). Military officers

 

”8 As would be expected the de facto powers of these political institutions was very limited and in fact the
personalities in charge and increasingly their relationship to Saddam Hussein dictated domestic political
actors relative importance.

113

were later to recount that the refusal to seize key Iranian cities was in part due to
Hussein’s micromanagement of the war (Wright 1985, 845). These failures led to an
almost immediate regression into a slow-paced attritional struggle for Iraq’s mechanized
forces.

Observing that a comparatively low cost-high benefits war was now unlikely
Hussein moved issue-goods position to one more amenable to a resolution of the conflict
(Chubin and Tripp 1988, 58).“9 The ante bellum issue-goods position for terminating the
war was restated intermittently throughout the first two years of the war, the preferred
issue-goods position of both the executive and most of Baghdad’s domestic political
actors (O’Ballance 1988, 77). Instead of reciprocating these offers Iran elected to
escalate the conflict, in 1981 it became Tehran’s espoused objective to topple the secular
Baathist regime.

To achieve Khomeini’s issue-good objectives a large-scale mobilization took
place. Employing superior numbers of soldiers infused with revolutionary zeal, the
Iranian’s introduced human wave tactics revealing both a new offensive capability and
great resolve (O’Ballance 1988, 61). The first half of 1982 witnessed a series of Iranian
victories that ejected the Iraqi army from Persian territory. The results of these battles
also produced a deepening sense of apprehension in Baghdad as that July the Iranians
began their invasion.

Hussein’s domination of military affairs was held to be a substantial reason for

the Iraqi forces’ cascading defeats in Iran. In organizing the defense the Iraqi executive

 

”9 “Nevertheless, the considerably reduced Iraqi peace conditions of 1982-in effect, indicating a
willingness to return to the status quo ante bellum — implied recognition that the nature of the war had
changed. In this there was an implicit admission that the war aims of 1980, once regarded as crucial to the
maintenance and extension of Saddam Hussein’s authority had also perforce to be abandoned [emphasis
original].” (Chubin and Tripp 1988, 58).

114

continued to micromanage the conduct of the war, exacerbating an already dangerous
military situation. The inefficient use of military resources appeared to be increasing the
hazard to the Baathist regime if Iranian forces were allowed to catalyze a mass Shiite
rally to Khomeini’s banner (Tripp 1993, 104-105; O’Ballance 1988, 88-89). Domestic
political actors began to assess Hussein’s war policies negatively.120

As the expected benefits of fighting declined for some Iraqi elites, political
pressure mounted to alter issue-goods position in order to prevent a collapse of the
regime. In one case a cabinet minister actually directly advocated the executive’s
removal in order to facilitate a negotiated peace. 12’ Such a bold or foolish effort
indicated a substantial level of political pressure that was building upon Hussein. While
the executive had a fairly strong capacity for elite repression, the war unleashed new
levels of repression upon his constituents. Purges and executions of military officers
were followed by the removal of high-ranking Baath officials (O’Ballance 1988, 91;

Karsh and Rautsi 1991, loo-167).122

 

'20 Some have disputed the likelihood of a Shiite revolt or believed that Hussein had co-opted some of these
tensions by elevating Shiites to positions of influence in his regime (Karsh and Rautsi 1991, 168-169).

’2' “The effect of war on the first grouping, the core of insiders, has been to underline their common plight.
Having reduced their numbers to sufficiently manageable and trustworthy proportions, Saddam Hussein
could rely upon them to endorse his view of the way Iraq should be governed, as well as of the function of
war in its future governance. . .Others, who believed themselves to be safe within the inner circle, have
discovered to their cost that being outspoken on the subject of Saddam Hussein’s own position can cost
them their careers and even their lives. The latter appears to have been the fate of the late Minister of
Health Riyadh Ibrahim Hussein, executed in October 1982. He had reportedly suggested that Saddam
Hussein might indeed relinquish the Presidency in order to comply with Iran’s major condition for ending
the war.” (Chubin and Tripp 1989, 88)

122 In a June 1982 meeting of Bath Party elites, eight of sixteen members of the RCC were removed,
seven were also scrubbed from the Party’s Regional Command, eight members of the cabinet were
replaced, and 300 high-ranking military officers were executed and a greater number purged. (Karsh and
Rautsi 1991, 166-167).

115

The number of domestic political actors was contracted and modified in order to
garner greater support for the executive’s policies (Chubin and Tripp 1988, 85).123 These
efforts engendered a more personalistic dictatorship where elites continued in their
positions of influence with the consent of Hussein. The role of the state was also
expanded in Iraqi society, including the creation of a cult of personality surrounding the

executive (Tripp 1993, 97-103).

Figure 7: Distribution of Elite Opinion in Iraq

Beginning of War After 1982-83 Purges
Executive’s policy Distribution of
preferences/ Distribution of Domestic Political
current war policies Domestic Political Actors’ preferences
\b Actors’ preferences

Executive’s policy
preferences/
current war policies

 

 

 

 

 

 

< e f f
Less favorable More favorable Less favorable More favorable
relative “tlllty relative utility relative utility relative utility

 

’23 “This has been the governing principle of Saddam Hussein’s relations with the Ba’th, the armed forces,
the economic entrepreneurs, and even, as was demonstrated in 1983, with members of his own clan.
Precisely to avoid becoming beholden to any one of these groupings, despite the demands he must make of
them during the war, Saddam Hussein has sought to play one off against the other. He has picked out
particular individuals among them for special favor or disgrace, and has emphasized thereby their degree of
dependence on himself alone and on his own authority. Destructive as this may be of any notion of
community within Iraqi political society, it has greatly facilitated the construction of a thoroughgoing
autocracy, focused on Saddam Hussein. Ironically, despite the fact that such a form of government is
precarious at the best of times, the war, at least since 1982, appears to have facilitated Saddam Hussein’s
ambition of extending the circle of his personal authority into all sectors of Iraqi life.” (Chubin and Tripp
1988,85)

116

A more centralized Iraqi dictatorship meant that the executive benefitted from
domestic political actors codependency upon his continuing rule, mitigating some of the
most urgent fears of deselection. This was especially true given that the Baathist elites
realized a high likelihood of punishment if Iran was successful in deposing the regime
(Chubin and Tripp 1988, 91). Still the executive’s refusal to consider resignation or
voluntary regime change meant that no end to the conflict was in sight.

In addition to the extensive repression of elites, strategy was also changed in
response to political pressure. Under duress by his constituents Hussein delegated greater
autonomy to his military commanders and reduced his micromanagement, permitting for
a more successful defense of Iraq (Chubin and Tripp 1988, 116-118; Cordesman and
Hashim 1997, 51).

Tehran looked to take advantage of its three-to-one advantage in population.
While at a disadvantage in terms of equipment Iran sought to inflict as many casualties as
possible upon Iraq to attrite their opponent’s manpower and grind down the population’s
resolve (Tripp 1989, 66-68). Defending against an Iranian invasion demanded massive
mobilization of resources which in turn increased the prospects of an internal revolt
through an exhaustion of the public (O’Ballance 1988, 98). If the broader Iraqi populace
refused to mobilize or began a large-scale revolt would likely entail the same issue-goods
benefits as a military victory.

Hussein would embrace military strategies designed to minimize their own
casualties while maximizing their opponent’s, culminating in use of chemical weapons

(O’Ballance 1988, 149-150). Having settled upon a negotiation position of the pre-war

117

distribution of territory, the military strategy was a static attritional defense using
obstacles and fortifications in order to reduce losses.

The Iraqi’s defensive strategy managed to parry thrusts attempting to sever the
lines of communication between Baghdad and Basara. During the second half of 1982,
the Iraqi’s successfirlly blunted the Iranian offensive. The positive developments from
the battlefield helped increase domestic political actors’ perceived likelihood of staving
off regime change (Chubin and Tripp 1988, 58-59).’24

The steady military pressure being applied to the Iraqi army at was responded to
by a defensive attritional strategy. The difficulty with having assumed a passive defense,
in order to avoid casualties, was that it was very difficult to coerce the sought after status
quo settlement from Iran. Beginning in 1983 punishment strategy was employed,
attempting to choke off Iran’s ability to export oil (Cordesman 1987, 66) The ensuing
tanker war would continue intermittently throughout the duration of the conflict, but this
effort was only partially successful in depriving Iran of its oil revenues (Chubin and
Tripp 1988, 134-137).

Another manifestation of this punishment strategy became known as “war of the
cities”. These intermittent campaigns were conducted principally using surface to surface
missiles launched at major population centers. While causing panic amongst the citizens
of Tehran, the resolve of Khomeini was not broken on the account of targeting civilians

(O’Ballance 1988, 153-154).

 

'2‘ “Nevertheless, the thoroughness of Iraq’s defensive preparations, and the success with which the Iraqi

forces were able to throw back the Iranian offensive during 1982, seem to have put new heart into the
leadership and even encouraged them to believe that the war might be brought to an end by this means.
Iraqi soldiers were now exhorted by Saddam Hussein not to think only of defending Iraq but also of ending
the war. . .Some months later, following the successful defense of Basra, Sa’dum Hammadi [foreign
minister] echoed this by stating that the failure of recent Iranian offensives could be considered an end to
the war since it would cause morale to deteriorate in Iran, and enhance the desire of its leaders to end the
war. Skirmishes might continue, but the war was as good as over.” (Chubin and Tripp 1988, 58-59).

118

Iranian offensives in 1984 did not markedly move the front lines. Even when
Tehran was able to defeat the Iraqi’s in a sector of the front, the inability to exploit these
advantages by disrupting the Iraqi logistical system or have ready replacements for
devastated units preempted a major breakthrough (Coredesman 1987, 64-65). These
successes did not eliminate the specter of Iraqi resolve crumbling in a protracted war.
Fears of the destruction of the regime fed negative perceptions of earlier executive policy
choices particularly the decision to initiate the war (O’Ballance 1988, 47-49).
Additionally the restive Kurdish regions also continued to serve as a potentially
destabilizing factor for Baghdad’s writ in northern Iraq (Karsh and Rautsi 1991, 168-
169)

Despite effective efforts at cowing domestic political actors through elite
repression, the Iraqi executive would face increasing political pressure. The loss of the
Fao Peninsula in February 1986 was a stunning blow for the regime. Deprived of much
of Iraq’s proven oil reserves it appeared as if Khomeini had finally achieved a significant
breakthrough in the war of attrition. In response to this military disaster Hussein took
hold of the reigns of the military yet again and proceeded to order a quick counterattack.
The endeavor was extremely costly, especially given the reticence for casualties amongst
the executive and domestic political actors.

After this defeat Hussein’s personal prestige continued to be diminished and
throughout 1986 he sought to compensate for the disaster at Fao by seizing a substantial
piece of Iranian territory at Mehran (Chubin and Tripp 1988, 64-65). This set off another

costly enterprise netting no territorial gains. As with 1982, it appeared that the Iraqi

119

executive’s insistence on managing strategy and personalizing the war with Khomeini
was diminishing the perceived utility of further fighting.

Political pressure emerged as micromanagement of the military campaign and
demands for poorly conceived offensives appeared to be eroding Iraq’s military
capabilities. The lingering disquiet of what heavy casualties could mean for internal
stability also rattled Baghdad’s elites (Chubin and Tripp 1988, 89).125 The Iraqi
executive responded with another round repression to quash threats of deselection by
intensifying the patrimonial nature of Bath Party politics.

As the battlefield indicated that the likelihood of ensuring the regime’s survival
declined and the utility of the war decreased for domestic political actors, the executive
again modified military strategy. Following the loss of Fao and the defeat at Mehran
Hussein began to loosen his control over military affairs. While certainly not suffering
that any potential rivals emerge in the military leadership, a more professional approach
to strategy was employed (Tripp 1989, 71).

As the war continued Iran’s revolution stoked fear amongst the Persian Gulf

states. Wariness of Shiite unrest was exacerbated by Iranian attacks upon these Arab

 

'25 “The two years in which these rifts within the regime became apparent—I982 and 1986 —were years in
which the military and strategic judgment of Saddam Hussein had been demonstrated by Iranian military
success to have been wholly misguided. As a consequence, his authority as the paramount leader had
projected himself to be came into question, or at least be sensed that this might be the case. Just as in 1979,
when he sensed a similar crystallizing of opposition to his leadership within the senior levels of the RCC
[Revolutionary Command Council], government and party, reacted ruthlessly to destroy it. He thus
simultaneously narrowed the circle of those of whom he could trust politically, and went to considerable
lengths to emphasize his own personal authority and indispensability.” (Chubin and Tripp 1988, 89).
“The same sequence of events appears to have taken place in 1986, when it became clear that Saddam
Hussein’s attempt to compensate for the loss of the Fao peninsula by authorizing the ill-considered capture
of Mehran had ended in disaster. Although this did not lead to the purges and dismissals on the scale
witnessed in 1982, it did lead to a noticeable tightening of Saddam Hussein’s personal control within the
RCC and the Ba’th, in part by placing yet more relatives and protégés in senior positions.” (Chubin and
Tripp 1988, 92).

120

states’ interests. Tacit support for Iraq resulted from instability in the region.’26 The US
provided satellite intelligence, arms from Egypt, financial support from Saudi Arabia,
were among the contributions that helped permit the Iraqi army the resources to again
take the military initiative (Chubin and Tripp 1988, 214-216).

Having reconsolidated his authority the Iraqi executive was able to launch a series
of attacks in 1987-88 that forced Iran from all their holdings in Iraq. Facing a revitalized
opponent in Baghdad and becoming increasingly isolated internationally Tehran agreed
to Hussein’s offer of an ante bellum settlement.

The Iraqi executive faced intense political pressure at different points in the war
as the threat to the regime grew, the intensity of which paralleled that of Iran toward the
end of their war and exceeded that which Mao faced. Despite enjoying a superior
capacity to repress domestic political actors than nearly any other executive studied here,
as the threat to the regime increased Hussein was forced to make accommodations to elite
concerns.

Elements of war policies were broadly supported by both the Hussein and his
constituents such as the ante bellum status quo negotiation position. With the emergence
of political pressure by intense repression of elites was undertaken. Additionally military
strategy was altered with greater control delegated to more skilled, if still politically
reliable officers. The improved prospects for the survival of the regime mitigated much

of the political pressure amongst Iraqi elites. However, the unitary assumption of

 

’26 Eventually the ability of Iraq to gain access to supplies of weapons and sympathy from the West and the
Gulf States permitted Hussein a more robust ability to attack Iran’s oil economy than they could be
damaged in return. This low-cost alternative to concluding the war through exhaustion was designed to
reduce casualties, but did not succeed in bringing Iran to a settlement for years. Even as the United States
tacitly defended Arab interests and permitted attacks on Iranian ships and oil platforms. (Chubin and Tripp
1988, 214-216).

121

principal of convergence model also fails to account for the changes in war policy
brought about by the application of political pressure.

The importance of casualties and other losses in terms of catalyzing changes in
tactics is also a possibility, as Iraqi military posture moved to minimize Iraqi casualties.
But the change in strategy occurred only after it appeared as political pressure was being
applied to the executive. Making an objective analysis of key variables an insufficient
explanation for why changes in strategy occurred.

The patrimonial nature of Iraqi politics and the intense repression that could be
meted out against elites meant that Hussein could largely pursue his preferred war
policies. Still this did not mean that the executive’s unitary preferences dominated all
other considerations. Political pressure emerged after battlefield defeats illustrated a
diminishing ability of the Baathist regime to defend itself. The blame for military defeats
in 1982 and 1986 were attributed to Hussein’s micromanagement and poorly
conceptualized offensives. Increased casualties and enhanced the likelihood of imposed
regime change or internal revolt engendered political pressure which was responded to by

alteration in strategy.

Soviet Union: World War II

The Soviet Union’s executive managed one of the most completely totalitarian
societies in history, yet the invasion of Stalin’s massive domain would challenge even his
ability to repress dissent. Mitigating political pressure necessitated elite repression as

well as adjustments in military strategy.

122

Stalin’s capacity to quash elite dissent had served the executive well. The
repressive apparatus truly terrorized both domestic political actors and the broader
populace. Denunciation of high-level communist party officials before and during the
war was commonplace (Rayfield 2004; Borekmeyer 2004).

Domestic political actors consisted of the upper echelons of the Communist Party
represented by the Politburo and their deputies, general officers in the Red Army, and
similar ranking leaders of the repressive apparatus the NKVD (People's Commissariat for
State Security forenmner of the KGB). These elites privy to an understanding of what
had occurred in the days following the German invasion as well as the willful self-denial
before the assualt and knew fault was to be found principally with the Soviet executive.

The Red Army was largely unprepared for Operation Barbarossa (the German
invasion of the USSR), with Stalin refusing to accept what any number of intelligence
sources were telling him about German plans.‘27 By 22 June 1941, Stalin had ordered
basic mobilization, but he also demanded extreme caution in any preparations along the
border not wanting to provoke Hitler (Rotundo 1989, 293-294). The devastation inflicted
upon the Soviet armed forces in the opening weeks quickly imperiled the regime.

The officer corps of the Red Army had been devastated in the years leading up to
World War 11 through mass purges. In the summer of 1941 three-quarters of all Soviet
officers had been in their current positions for less than one year (Bialer 1969, 63). And
much of the confusion of the early days of the war is attributed to inexperience of the

military leadership (Erickson 1975a, 136-138).

 

'27 It has been suggested that the Soviets were prepared more than is generally claimed, in fact he goes as
far as to suggest that the German rationale for the war was somewhat justified by the large Soviet military
presence on border (Magenheimer 1999, 51-59).

123

Coming to grips with his own folly, a sullen executive thought that he was going
to be deposed as the Soviet forward forces were annihilated (Derek 2005, 191).128
However upon being accosted by some of the most influential domestic political actors in
the USSR, these elites indicated their willingness to retain Stalin. Bereft of any other
mechanism to mobilize the Soviet leviathan state, Molotov (foreign minister) and Beria
(head of the NKVD) amongst others sided with the executive rather than hazard the fight
without him (Brackman 2001, 349).

Having been assured of his position in the short term, Stalin set about conducting
the war as he saw fit. In the months immediately following the 22 June 1941 invasion,
the executive’s centralized control allowed for rapid mobilization at the cost of the
executive’s micromanagement of military affairs (Glantz 2005, 68).’29 Interference by
the Soviet executive into military planning brought further disasters upon the Red Army

(Erickson 1975a, 172-175; Bailer 1969, 34-39).

 

'28 “On 30 June, Molotov [foreign minister] summoned Mikoyan [Politburo member] and Voznesenski
[central economic planner] by telephone. They found Malenkov [politburo candidate], Voroshilov [Red
Army Marshall] and Beriya [head of the People's Commissariat for State Security] with Molotov. Beriya
proposed the formation of a State Defense Committee, and it was agreed to ask Stalin to head the body
because of his position and ‘reputation with the people’. On Voznesenski’s initiative, Molotov was
delegated to lead the deputation to Stalin, the others promised their support and said that if Stalin was unfit
to lead the body they would support Molotov. . .When the group arrived at Stalin’s dacha, he appeared to be
worried, perhaps thinking they had come to arrest him because of his refusal to recognize the imminent
German attack in early 1941. Molotov, as spokesman, talked of the need to create a GKO [State Defense
Committee] to concentrate power and put the country on a war footing. Stalin asked who was to lead this
body, and Molotov said that Stalin would. Stalin looked astonished, but then agreed.” (Derek 2005,191).
’29 “Stalin’s demands that the Red Army cling to untenable positions and his meddling in strategic and
operational decision-making were directly responsible for its disasters at Uman’, Kiev, Viaz’ma, and
elsewhere in 1941. His influence deprived the Stavka[Main Command of the Armed Forces of the USSR]
of its initiative and limited its strategic horizons, forcing it to plan reactively to the single imperative of
restoring stability to the front, but his single-minded insistence on marshaling reserves and his ruthless, but
often stingy, allocation of these reserves ultimately strengthened the Red Army strategically. As a result,
the Red Army’s stubborn resistance during the battles for Leningrad, Moscow and Rostov and the energy,
sacrifice, and determination it displayed during the ensuing winter campaign reflected Stalin’s iron will.
Stalin’s strategic blunders notwithstanding, in December 1941 the threadbare Red Army fought with a
ferocity and desperation that mirrored its leader’s determination and ruthlessness.” (Glantz 2005, 68).

124

Stalin demanded that immediate counteroffensives be undertaken to blunt the
Wehrmacht’s advance (Erickson 1975a, 85). Prefabricated war plans led to ill-prepared
and under-strengthened assaults against Hitler’s forces (Rutondo 1989, 286-288). These
piecemeal attacks squandered vital resources and fed into the larger problem of Red
Army units engaging the Germans at the front only to be enveloped and cutoff by swift
panzer penetrations behind them.

So drained of autonomy by mass purges, even in the face of monstrous orders
senior officers quailed rather than voice criticism to the executive.'30 When the Soviet
executive ordered that 700,000 soldiers hold their positions around Kiev rather than
retreat only minimal perfunctory protest was heard from a single general (Bialer 1969,
38). Tens of thousands of Red Army soldiers were captured soon thereafier. By the fall
of 1941 the Red Army had lost over 2,800,000 soldiers killed, wounded, or captured
(Rayfield 2004, 401).

The shock of Barbarossa and the Wehrmacht’s well-practiced blitzkrieg strategy
had precipitated an epidemic of surrender and panic. In response Stalin installed
draconian punishments with the NKVD extermination battalions or SMERSH, ordered to
execute those found guilty retreating or deserting (Magenheimer 1999, 95-97). The
coercive mechanisms of the Soviet state did help to arrest a collapse of morale, but the
rigidity of the executive’s control undermined the likelihood of preserving the regime.
This engendered intensifying political pressure even in the face of elite repression as
domestic political actors would have both their positions and their lives imperiled by a

German victory.

 

'3° “Despite the early military setbacks and obvious role that Stalin played in the disaster. He still was able
to dispose of high-ranking communist party members with impunity, a frequent occurrence in the early part
of the war with Germany.” (Glantz 2005, 399).

125

Doubtless, Soviet elites ascribed a very low probability of achieving even an ante
bellum status quo settlement during the early part of the war. Two of the highest ranking
party members, Beria and Molotov, were both said to have endorsed a major territorial
concession similar to the Brest-Livtosk treaty in the fall of 1941 in order to garner a
peace with Hitler (Brackman 2001).131 The reticent executive was not in the mood to
compromise with his antagonist, especially after having been duped by the Russo-
Gerrnan non-aggression pact. Unwilling to move his issue-goods position to one in
which Hitler was likely to accept, Stalin continued to use elite repression in order
mitigate the threat of deselection.

The executive was indispensible to the highest echelons of the communist party
because of the ability to coerce compliance and the popular identification that Stalin had
with the USSR. Domestic political actors acquiesced to his leadership in the belief that
he was the best option for a quick mobilization. The faith placed in the executive’s
ability to catalyze Soviet mobilization was partially vindicated.

Operation Typhoon commenced in the autumn of 1941 and was the German
command’s last effort to finish the war. It appeared that the Wehrmacht would produce
another miraculous victory. But the Red Army reinforced with fresh divisions from
Siberia prepared for the inhuman cold launched a series of counterattacks that forced the
Germans back from Moscow in the winter of 1941-42. Due to logistical exhaustion as a
much as from the defensive prowess of the Soviets, the winter had settled in without the

capitulation of the capitol. Surviving the first onslaught did not mean that the regime was

 

’3' While some early scholars suggest that Stalin helped to orchestrate some peace overtures to Hitler from
the Summer of 1941 until the Fall of 1943 (Erickson 1975b, 162), other scholars refute this possibility
(Roberts 2007, 167; Vojtech 1972). The incredulous scholars suggest that Stalin having been burned by
Hitler’s offers before would not find future German offers credible. Further, these authors also cite the lack
of firm documentary or other historical evidence for sanctioned negotiating offers being made by Moscow.

126

secured and Stalin’s leadership appeared to be reducing the likelihood of preserving the
regime.

Whereas Hitler had demonstrated some strategic genius in the early part of the
war (Mearsheimer 1999), that was not Stalin’s strength (Bialer 1969, 43).132 The German
executive also had a much more established institution in the Wehrmacht officer corps
than did his antagonist. Moreover Stalin’s mass purges had so cowed his senior officers
that autonomy and competence were much reduced. With few professional soldiers
willing to admonish executive strategy or have their advice heeded, even in the midst of
the USSR’s greatest peril the efficiency with which military assets were used was
abysmally low.

In the spring of 1942 the Red Army was again ordered to make hasty attempts to
wrest the initiative from the Wehrmacht. These overly-ambitious offensives squandered
the potency of any one attack and led to still greater Soviet losses, as illustrated in the
following passage.

Stalin’s misjudgrnents also contributed to the Red Army disasters of Khar’kov and the Crimea in
May 1942 and the cascading series of defeats it experienced on the road to Stalingrad in the
summer and early fall of 1942: “The chief reason for the failure of the summer campaign of 1942
was the High Command’s erroneous decision ‘to aflix’ numerous separate offensive operations on
all fronts to the strategic defensive operation. This dispersal of strength and premature
expenditure of strategic reserves certainly doomed Stalin’s plan to failure.” Glatz 2005, 68

 

’32 “While Hitler had less control and power over the military than Stalin, he relied less on the military than
Stalin. Of the many facets of wartime leadership, ranging from the symbolic to the administrative, Hitler
was increasingly drawn to military planning and operational command. With the progress of the war his
contempt for and isolation from his generals, his own initiative and judgment concerning military
operations grew until it became in the last stages of the war the dominant aspect of his leadership. An
inordinate self-confidence fed on the exhilaration of early victories achieved over the skepticism and
protest of his military professionals. Stalin, as the war memoirs show, never considered the conception and
planning of military operations his paramount strength, his major interest, or the measure of his absolute
authority. He was more willing to listen to his generals, more willing to correct his errors. . .This crucial
difference in the two leaders’ respective attitudes toward their military associates enabled Stalin to benefit
more than Hitler from the improvement in professional quality of officer cadres which both sides witnessed
over the course of the war with experience and the promotion of talented commanders. Hiter’s generals
exercised less influence on the decisions of their High Command at the moment they were most able to act
effectively. Stalin’s generals exercised more.” (Bialer 1969, 42-43)

127

In the spring of 1942 Stalin’s continued control over military affairs led to more disasters
reducing the likelihood of securing the regime. Whereas the executive’s ability to
mobilize rapidly and vindictive enforcement of his commands had reconstituted the
state’s defenses, now it appeared that Stalin was a serious liability upon the war effort.

Domestic political actors believed that defeat could mean their positions of
influence and their lives lost if the state should succumb to the German onslaught.
Political pressure regained intensity as the spring and summer of 1942 witnessed even
more military debacles.

As costs increased in terms of military assets the likelihood of preserving the
Soviet Union diminished benefits, demands grew for an alteration in war policies.
Paranoid to the point of observing conspiracy everywhere, Stalin had a highly developed
capacity to intimidate potential rivals. But this did not mean that deselection was not
possible as accounts of Stalin’s behavior following the initial German invasion indicated
(Brackman 2001, 369; Derek 2005, 191). With the military position of the USSR
worsening Stalin began to heed the desperate lobbying of domestic political actors and
permitted military elites to define strategy (Glatz 2005, 68-69; Bailer 1969, 341-342).133

With reduced interference fi'om Stalin, the Red Army absorbed the summer 1942
German drive upon the Volga River. Instead of squandering military efforts in hasty
counterattacks, the Soviets were eventually able to direct large numbers of reserves into

the Stalingrad area. After fierce resistance in the metropolis itself, the Soviets launched

 

’33 “At Stalingrad in the fall of 1942, however, Stalin replicated his positive performance of the previous

year-but only because he began to heed the advice of his most trusted key military advisers, such as
Zhukov, Vasilevsky, Antonov, and Voronov. Thereafter, Stalin continued following his advisers’ counsel
throughout the remained of 1942 and all of 1943, though not without retaining tight control over all of his
political and military subordinates. As he had earlier in the war, when he deemed it necessary, he treated
those he suspected of disloyalty with harsh disciplinary measures, and he often confused combat failures or
perceived ineptitude on the part of field commanders with outright disloyalty.” (Glatz 2005, 68-69).

128

Operation Uranus in the winter of 1942-43 that shattered the poorly defended flanks of
the German 6th Army. The envelopment of the 250,000 soldiers in Stalingrad and
permitted the Soviet military successes in the following months as the Wehrmacht was
driven from the Caucuses (Erickson 1975a; Erickson 1975b).

Justifiably considered the turning of the tide on the Eastern Front, Stalingrad
signified that the mortal danger to the Soviet Union was at an end. From the perspective
of domestic political actors these victories indicated a reduction in cost by using military
assets more efficiently. The alteration from Stalin’s unitary appraisal toward one in line
with Red Army professionals helped to reduce the likelihood of a regime’s destruction
and increase the likely benefits of further fighting.

In summer of 1943 the Wehrmacht had again recovered, but the balance of
military power had shifted markedly against Germany. Looking for a victory that could
secure some negotiated settlement Hitler ordered a major offensive against the Soviet
salient around Kursk (Magenheimer 1999, 198-201). Stalin had wanted to launch a
preemptive attack against the German lines, but was eventually dissuaded by his generals
(Erickson 1975b, 68-75). The Red Army instead dug in with redundant defensive lines
and was able to do permanent damage to panzer formations. Afterward the Red Army
never relinquished the initiative driving the battered Germans inexorably toward Berlin.

Changes in military strategy did not appear to be inspired by some sensitivity to
human or other losses in fact casualty aversion would be one of the last descriptions for
Soviet strategy in World War II. A risk acceptant strategy employed in order to secure

the executive against punishment following the war is also not relevant to the Soviet case

129

given the highly capable ability to repress the populace. Furthermore a status quo
settlement likely would not have meant the deselection of Stalin.

The level of political pressure in the opening weeks of the Soviet-German war
rivaled the intensity of any other episode in the other autocratic case studies. Political
pressure was greeted with intense repression of elites and also mitigated by the
acknowledgement of many domestic political actors of the indispensability of the
executive for mobilization. But in the spring-summer of 1942 when the Red Army was
largely mobilized, but without an imminent winter to save the USSR, political pressure
again intensified.

Whereas issue-goods negotiation position was not seriously considered as a
means of improving the utility of domestic political actors, alteration of military strategy
became the subject of determined lobbying. The intensity of political pressure for change
in war policy grew, while elite admonishments of strategy during Barbarossa were muted,
lobbying for policy change occurred in the months following. In one of the most
repressive regimes in history the executive was not able to dictate his preferred war

policies without modification.

Conclusion

The above chapter illustrates the key concepts of the theory and then compares
them to alternative hypotheses of why war policies might change. In all cases some
dissent with executive war policies is noted. A difference within and between cases is the
proportion of domestic political actors sufficiently dissatisfied with war policies to hazard

an attempt to deselect the executive. Inherent in political pressure driving deselection is

130

the level of repression that the executive can wield against elites in an autocracy.
Doubtless an executive in an autocracy enjoys greater latitude if they can coerce elite
dissent.

In response to either acts of political pressure or an appreciation of the latent
opinions of domestic political actors, repression is one of the principal mechanisms by
which elites stay in power. The obvious appeal of repression is that it allows for the
executive to pursue his/her preferred policies even in the face of political pressure. As
was the case with Le Duan’s use of repression in order to execute his preference for
major unit combat in South Vietnam. Additionally the size and composition of domestic
political actors can be altered to help secure a more compliant constituency in the future,
as was the case in post-Revolutionary Iran.

An executive’s ability to repress elites acts as a deterrent in terms of the
likelihood of hazarding a coup or assassination attempt. The collective action problem
would suggest that the creation of a hazardous conspiracy with a small number of elites
likely reflects the sentiments of a larger number of domestic political actors. The July
1944 assassination plot against Hitler was conducted by a relatively small number of
military officers, but reflected the sentiments of a larger cohort of domestic political
actors.

More hazardous or violent manifestations of political pressure likely reflect a
substantial portion of a state’s elites share a comparatively negative view of war policies,
if not necessarily on what policy alternatives should be implemented. But because the
decision rules for deselection of an executive are not established a group consisting of

less than a plurality of domestic political actors could usurp the leader.

131

Executives with a greater capacity to repress elites have a somewhat greater
ability to pursue their preferred war policies, but the ability of a state to mobilize can also
be degraded by quashing innovation and reducing competency. Stalin’s purges of the
Red Army before and during the war were a major contributor to the defeats suffered in
the weeks following the German invasion. Even when accounting for the ability to
repress one’s principals, extensive use of elite repression can undermine the regime’s
security.

The Japanese case offers an extreme in this respect. Intense political pressure is
registered as it became clear that major defeat at the hands of the United States was a
likely possibility. Having little ability to repress elites, Tojo refused to countenance a
serious negotiated settlement or to substantially alter military strategy consequently he
was deselected. Even in less extreme cases, where a mortal threat to the regime is less
ominous and the executive has greater faculties to coerce elite dissent similar findings are
observed.

Chairman Mao was clearly a much more dominating figure than was Prime
Minister Tojo and the threat to the People’s Republic was much less severe, yet the
Chinese executive was still dissuaded from his preferred course of action by the
possibility that he might experience a reduction in authority. Although an earlier
reticence to alter war policies was present, political pressure gradually built as other
facets of the regime’s security appeared to be endangered by the Korean War.

Post-revolutionary Iran witnessed a marked shift in the relative power of the
executive, as Khomeini consolidated authority by eliminating domestic political actors

with whom he had limited ideological affinity. However, later in the war as an existential

132

threat to the revolution loomed the reduced class of elites in the regime applied political
pressure eventually succeeding in ending the war.

These autocracies with a limited capacity to repress elites all illustrate the
interplay between political pressure and the imperative to alter war policies. It would
appear that assessments of a war’s utility are made well before the conclusion of a
conflict. If domestic political actors also fear punishment that would come at the hands
of a popular revolt or a foreign invasion, they would likely attempt to alter war policies to
reduce these threats.

In North Vietnam it appeared that the executive was supported by a large portion
of domestic political actors in terms of how the war would be conducted. Given a modest
amount of repressive ability, the executive was able to pursue an aggressive strategy to
reunify the country. However, Le Duan was forced back away from his preferred
strategy on several occasions as the utility of current war policies was being negatively
assessed. Whereas the threat to the regime was moderate, the lingering threat of an
American invasion or a crippled economy helped engender political pressure.
Consequently the attritional strategies and reticence for meaningful negotiations were
dropped in favor of less costly policies.

The dictatorial Nazi state with its powerful executive was not immune to political
pressure. Hitler injected himself into all manner of military decisions and as Germany’s
fortunes declined, demands for policy changes were made. Lobbying for policy changes
did not garner the intended effect, though occasionally the executive would yield to the
advice of professional officers. The unresponsiveness to political pressure brought about

several attempts to deselect Hitler as well as changes in strategy designed to secure his

133

position. While repression managed to secure the executive throughout much of the war
with only modest adjustments to his preferred policies, the refusal to alleviate political
pressure nearly resulted in his deseletion on several occasions.

Hussein fared better than did the German executive surviving the conflict with the
regime intact. While the threat to Iraq in its war with Iran was certainly less desperate
than was Hitler’s plight in the latter part of the war, Hussein also mitigated political
pressure through changes in policy. The negative assessments of war policies in the
midst of two different military crises led to increased levels of repression and alterations
in strategy designed to improve the utility of his constituents. This case offers yet
another example of how even leaders with impressive elite repression capabilities are
compelled to alter their preferred policies in order to ensure their political survival.

The highly dictatorial state observed similar outcomes. Stalin feared for his
position as the threat to the Soviet Union became evident in the opening weeks of the
German invasion. War policies were altered in response to political pressure, at the war’s
onset and again as it appeared that executive micromanagement was undermining the
likelihood of the Soviet Union’s survival. As the threat to the regime abated the principal
impetus for political pressure also subsided, though the executive persisted with a more
hands-off approach.

A number of inferences are drawn from these case studies. Domestic political
actors do not appear to forestall judgment regarding a war’s utility until the very end of a
war. The same updates with respect to resolve and capabilities that are theorized to serve
as an objective criteria by which states eventually conclude wars, do not appear to be the

sole purview of the executive. Nor do these updates in beliefs appear to be objectively

134

interpreted by all observers in a unified fashion, even in societies with a small number of
enfranchised citizens.

Consequently dissent with current war policies can take the form of lobbying for
alternative ways to conduct the war be it strategy and issue-goods negotiation position.
To the extent that the level of mobilized resources is a subject of contention between
executive and domestic political actors it is endogenous to the military effort needed to
attain the issue-goods objectives at stake.

The principal evidence for the political pressure theory is significant changes in
war policy. In the above analysis alternative hypotheses for these alterations in strategy
and issue-goods position were also offered. A number of examples of an essentially
unitary assessment of the costs and benefits of fighting are objectively assessed in an
autocracy.

A state may be forced into changes in strategy and negotiation position simply
because other options are unavailable as material resources or battlefield realities have
substantially changed. The German failure to execute a maneuver strategy later in the
war was in part a function of reduced relative capabilities or the devastation of
communist units in the Tet Offensive necessitated a reversion back to guerilla warfare.
Issue-goods negotiation position might also reflect an essentially unified understanding as
to the most favorable settlement. As was the case in the Iraqi decision to pursue a status
quo ante bellum settlement following the stalled assault on Iran or Hanoi’s monolithic
determination to reunify Vietnam.

Assuming that a diversity of interests with respect to war policies exists between

agent and principals, the selection of a war is held to be the principal means by which

135

domestic political interests affect war policy. The choice to fight a given war has been
shown to be related to a cost-benefit assessment of made by a state (Bueno de Mesquita
1980) and as an effect of domestic political considerations (Reiter and Stam 2002).

Referencing selection as a substantial determinant of war outcome has empirical
salience. The importance of selection as a determinant of war outcome implies that a
post-hoc assessment as to the outcome of a war is the point at which a review of
executive choices is made. If a war is selected and fought in a way that is harmful to the
interests of an executive’s constituency, this invites the possibility of deselection.

The foundation of executive accountability rests upon a post-hoc review of
policies, which appears to be incongruent with the empirical evidence. While an
effective review of war policies might take place after a conflict in shorter or less
significant conflicts, this does not explicate the imperative of autocratic elites to protect
their interests. Well before a war actually concludes assessments of a war’s utility is
made, anticipating the outcome of a war and the potential for a revolution or foreign
conquest.

Another implication for a post-hoe review of executive choices is the anticipated
punishment by the executive at the conclusion of a conflict. Semi-autocratic leaders who
lose a war are unable to repress either popular or elite opposition that now observes the
outcomes of an executive’s war selection. Theoretically this is supposed to lead to more
aggressive strategies termed gambling for resurrection (Goemans 2000). The Iranian
assaults into Iraq in 1986-87 or the Ardennes Offensive by Nazi Germany were designed
to reshape perceptions through a last gamble with a state’s diminishing military

I'BSOLII'CCS.

136

If a review of war policies is conducted by domestic political actors in a similar
fashion to the executive, then why would they allow their interests to be compromised by
a overly risky executive strategy? The more imminent threat to a leader’s position of
power would not be from foreign opposition at the end of a war, but from their
constituents. The above cases illustrate that the fear of punishment amongst domestic
political actors that instigates pressure to alter war policies to ones that are less risky or
more likely to cut the losses of the state. Instead of anticipating reaction to a post-hoe
review of a war and the need to buy off dissent, elites appear to assess a war well before
the conclusion of a conflict and these concerns have to be engaged by the executive.

Specific to strategy change, key variables such as shipping losses or body counts
do not serve as a catalyst for changes in military strategy (Gartner 1997). Instead it
appears to be the case that overly costly military strategies are more frequently redressed
when domestic political pressure is applied to the executive. Autocracies might be more
insulated from public dissent and therefore weight costs differently than do great power
democracies. However, the interests of autocratic elites are protected through the
application of political pressure.

Executives across different types of autocracies are subject to the potential for
sanction and the influences of political pressure. The behavior of even dictatorial
regimes such as Stalinist Russia and Baathist Iraq had executives who bowed to the
policy critiques of their constituents. Additionally, the model of intra-war assessment by
domestic political actors makes much more realistic assumptions of elite behavior in
autocracies. Why would the beneficiaries of private goods jeopardize their positions of

influence by suspending dissent with war policies until the conclusion of a conflict? The

137

more dubious assumption is that autocratic elites are unable to understand the
implications of war policies, but that the executive can forestall an assessment of
information to domestic political actors.

The next chapter begins a similar analysis as to the salience of political pressure
in democratic cases. The lack of a repressive capability to be employed against domestic
political actors means that political pressure should be in effect to a much greater extent
that is the case in autocracies. Moreover, the scope of fi'anchise should also mean that the
diversity of interests represented in democracies should make maintaining support for

war policies at least as difficult as it is in autocratic cases.

138

Chapter Four: Democratic Cases

As with the previous chapter, the following democratic cases are examined to see
if the political pressure theory conforms to the historical record. The assumption of an
intra-war review by domestic political actors has been explored by the casualties as costs
literature. The implication that public discontent with casualties can engender outcomes
other than electoral sanction and demands for demobilization has not been explored.
Instead the principal implication of the democracy and war is that the executive of a
democracy is very selective of which wars to fight. Should they choose poorly and a war
becomes increasingly costly or extends too long the public will compel the state to
demobilize.

Similarly, even if a war is not selected well this does not necessarily doom the
enterprise. A state might alter tactics in order to improve military efficiency (Gartner
1997). Similarly the benefits of fighting might be revealed as being greater than
previously expected as was the case when it became widely understood that Nazi
Germany was perpetrating the Holocaust.

It has been demonstrated in a number of studies the importance of casualties in
determining the level of support that a war enjoys amongst the public. What has not been
determined is the relative strength of the empirical balance between costs and benefits as
appraised by a democratic public. Furthermore if there is a substantive benefit offsetting
costs than is the relationship between war duration and battlefield events monotonically

negative in terms of its effect upon the public support for war mobilization.

139

The democratic cases differ from the autocratic cases in some respects. First,
democratic executives are more reliant upon changes in military policy instead of
repression in order to maintain support for mobilization. Second, institutions in a
democracy mean that the size of the enfranchised population is larger and potentially
more diverse in terms of the opinions regarding war policies. Third, a failure to
accommodate political pressure results in an easier and less hazardous process of
removing an executive.

Democratic executives often face the critiques to war policy emerging fi'om the
public that are bimodal in their distribution. Obviously this is a simplification of mass
opinion, but the public does not uniformly demand demobilization. Instead political
pressure emerges for demands for escalation or alternative employment of mobilized
resources. While there is no novelty in suggesting a dove-hawk dimensionality of public
opinion (Feavar and Gelpi 2004; Verba and Brody 1970), navigating the very different
critiques of war policy is a challenge for democratic executives.

The following cases include the United States’ highly salient objectives of World
War II. In this conflict political pressure does not emerge in a serious way, although
policy changes are made in order to bolster public support for mobilization. Cases of
lower issue-goods salience include the American wars in Korea and Vietnam. Both
conflicts witness executives who fail to address the concerns of their principals is
deselected. The new presidents replacing these leaders demonstrate a greater degree of
flexibility.

The asymmetries in information between an executive and the public in a

democracy versus an autocrat and his elite constituents might be considerably different.

140

Moreover the level of military expertise amongst the public is likely considerably less
than is the case amongst domestic political actors. However receiving signals from the
political opposition and with freedom of the press policy alternatives can be presented to
the public. Regardless I theorize that the unitary preferences of these leaders are laid
aside in order to accommodate mobilization and reduce the likelihood of deselection.

I also propose a level of nuance to the democratic public has a monotonic reaction
to casualties and war duration. The cost-benefit calculations of a democracy’s domestic
political actors can, in theory, improve on the basis of changes in war policy. Casualties
and war duration have to be contextualized by the benefits being obtained through
fighting. If costs can be reduced and benefits enhanced than a more broadly applicable

theory to intra-war behavior of states that transcends regime type can be formed.

United States: Vietnam War

It has been argued that the policy of steadily deepening involvement in Vietnam
was generally reflective of sentiments of public opinion channeled through the executive
branch (Gelb and Betts 1979). Johnson’s incremental escalation of the Vietnam War
produced growing costs with only modest increases in anticipated benefits. Conduct of
the war in Vietnam led to varying critiques of the administration’s policy, for both greater
escalation and for demobilization. As political pressure mounted the Johnson
administration did not accommodate public opinion with changes in war policy.

American involvement in Vietnam was deepened by the administration’s

assessment indicating the growing fragility of the Saigon regime (Longevall 2004, 197).

After the Tonkin Gulf incident and the affiliated resolution passed in August 1964

141

endorsed mobilization, low-intensity bombing was initiated over North Vietnam. The
bombing effort was expanded in March 1965 in what became known as the Rolling
Thunder campaign and in May US combat battalions were brought into the country in
order to protect American installations.

Throughout 1964 efforts to start negotiations went nowhere as the American
executive insisted upon maintaining Saigon’s sovereignty and its alliance with
Washington as the minimal acceptable issue-goods position. A formal public articulation
of the US position was made in April 1965. This offer was sweetened by including
economic assistances to the DRV in exchange for an independent South and a
discontinuation of Hanoi’s support for insurgency (Moise 2004, 75). While Johnson was
anxious to get negotiations started, the April 1965 offer was essentially immutable
(Kaiser 2004, 52). ’34 Sensing an equally uncompromising attitude in the Hanoi
leadership, Johnson elected to achieve a self-sustaining RVN through military victory and
then negotiate from this position of strength (Kaiser 2004).

Aside from the bombing campaign of the DRV, the principal tactics being
employed against the communist insurgency was an attritional strategy. The US Military
Assistance Command in Vietnam (MACV) was led by General William Westrnoreland.
Under his command the South Vietnamese Army (ARVN) principally focused on
defending urban areas and conducting lower-intensity combat operations. American
maneuver battalions then conducted sweeps of the rural areas attempting to engage large

VC and NVA formations and crush them through lavish use of firepower.

 

'34 “First, they meant no cease-fire, which the United States was determined to reject until it had a clear
military advantage and evidence of Hanoi’s willingness to meet American terms. Second, they meant no
immediate agreement to do anything but exchange views. Johnson had managed to express a willingness to
talk without indicating any willingness to negotiate, much less to abandon, any of his objectives [emphasis
original].” (Kaiser 2004, 52).

142

In order to secure the ends of an independent South Vietnam, the means employed
was to annihilate as many communist soldiers as possible. Eventually eliminating more
soldiers then could be replaced through communist recruitment in the South or infiltration
from the North. Upon reaching the “cross over point” the insurgency’s potency would be
progressively reduced (Gartner 1997, 127-140).135

These military goals were to be obtained through a massive mobilization. At the
end of 1965 184,000 American personnel were in the Vietnam theatre, with 1,369 troops
killed in action. By the end of 1967 these figures stood at 485,000 and 9,378 (Clodfelter
1995). As is widely understood, the growing human costs of fighting were shown to be a
major factor in declining support for the Vietnam War (Gartner, Segura, and Barratt
2004; Gartner and Segura 1998; Mueller 1973). But the relationship was not as simple as
the number of battle deaths lowering domestic support, instead the benefits of fighting
were used as justification for the sacrifices made. The Johnson administration attempted
to extol the public for further mobilization by trumpeting progress in the war (Gelpi,
Feaver, and Reflier 2005/2006; Blood 2006, 14.15)}36

The strategy of eliminating more communist soldiers than could be replaced
served as a metric of success for the military, but was also used to bolster public support
for mobilization. The intermediary statistic of body counts was deputized in order to

demonstrate incremental progress toward quelling the insurgency (Boettcher and Cobb

 

”5 The US strategy did produce heavy casualties amongst the communist soldiers, but these losses did not
consistently exceed the DRV and VC’s ability to replace eliminated soldiers.A CIA memo indicated that
even when accounting for rebuilding after bomb damage and the defense of the DRV, the crossover point
was not reached, this compared to comparatively optimistic assessments by the DIA as to the VC-NVA’s
troop strength in the South (Gartner 1997, 145).

Westmoreland believed with some validity that a crossover point had been achieved in April 1967,
however this did not mean the imminent demise of the communist war effort in the RVN (Blood 2005, 28-
29).

’36 The authors indicate that the president can rally the public by pointing to episodes of success and stating
the case for further mobilization (Gelpi, Feaver, and Reflier 2005/2006).

143

2006; Gartner 1995). Pointing to somewhat inflated body count figures and relying upon
dubious intelligence reports the American commander in Vietnam attempted to

demonstrate the benefits of fighting.

Figure 8: Lowess Graph of Public Support for Vietnam War and War Duration137

 

70 l
I
i Nov-65

60%i \‘xAug-65

Percent l \ \ Mar-66
Approval for l " .\
Vietnam War N Fegé67

50 J ‘\\\ 0V“

l May-66“\.\May'67
Sep-66 "xJul-67 Dec-67

I “~QCt-67
3 ‘ ~ «..feb-68

40 ] \\, Mar-68 F eb-69

\\~. Apr-68

l ~~~~~ Oct-68 May-70
l """"""" Apr-70 Jan-70
, Aug-68 T“ ........ R Mar-70

304 Sep-69 “rs-9.-.“, Jan-71

Apr-71
IL_l_ I I 1 I
0 20 40 60 80
War Duration (Months)

bandwidth = .8

The culmination of these efforts to rally the public took place at the National
Press Club on 21 November 1967 when Westmoreland stated that “We have reached an

important point when the end begins to come into view.”138 The benefits of the war were

 

’37 Data come from Mueller (1973) reprinted in F eavar and Gelpi (2004. 137)-
‘38 Quotation from Oberdorfer (1971, 105).

144

illustrated by a high-probability of achieving a sustainable RVN and achieving the
security goods of containing the spread of communism throughout Southeast Asia.

Efforts to enhance the public’s perceptions of the likelihood of achieving these
objectives were coupled with statements of the relative irnminence of achieving victory.
Onl 5 November 1967 Westrnoreland told reporters of the overwhelming optimism
throughout MACV and that if current progress continued the US would achieves its
objectives in two years (Oberdorfer 1971, 104-105). While extolling domestic political
actors’ to continue to support mobilization was done to offset the costs of fighting, efforts
that maintained support for the war but also introduced audience costs.

The January-February 1968 Tet Offensive offered evidence as to the surprising
strength of the VC in the RVN. The effect of the daring, and calamitous, attacks was to
credibly demonstrate a military capacity that exceeded the Johnson administration’s
portrayal of the communist forces. Research conducted since has found that these attacks
fundamentally altered public impressions of the war (Blood 2005, 21, 66-67; Buzzanco
1997, 231-249; Gelpi and Feaver 2004, 138-139).

Even at current levels of mobilization domestic political actors now drastically
reduced their impressions of the likelihood of achieving the issue goods in question as
well as increased the perceived time until the war could be won (Lunch and Sperlich
1979, 40). What came to be known as the “credibility gap” reflected the public’s
perceptions of the benefits of the war were substantially lower than that which the
executive had fashioned.

The high levels of mobilization demanded by Johnson’s prosecution of the war

could not be sustained as these high costs, especially in lives, were not offset by the

145

newly deflated appraisal of benefits. However the post-Tet decline in support for the war

implies a monolithic class of domestic political actors assessing the war and rendering

 

 

judgments.
Figure 9: Percentage of Public Supporting Escalation and Withdraw139
60 n
l
} / / ”\‘I‘SOVWC Supporting Escalation
//May-67
' F eb-67 \
40 i ~. Nov-70
Percent : /,'"'Nov-66 Mar—69 ’
Support ' ,/ Nov-68 \qu ,
l l/"Mar-69 Nov-70
. /F -
20 A, eb 66
l \ Feb-66 Nov-68
! \
\ - ,/
\No" 66 ,,/’Nov-6:\
\ \ / May-67
. F eb-67 Percentage Supporting Withdraw
o —-l
T "-" ‘ I I T—
20 40 60 80
War Duration (Months)

Hawks and doves articulated very different critiques of the executive’s war

policies. A survey in the spring of 1967 found that 18 percent of the population were

hawks and another 31 percent were mild hawks, 26 percent were considered to be in the

center with 13 percent mild doves and another 12 percent ardent doves (V erba and Brody

 

'39 Data comes from Lunch and Sperlich (1979, 27-28).

146

1970, 330). These different cohorts of domestic political actors generally became
progressively more dissatisfied with the conduct of the war, but responded with the
advocacy of different policy alternatives.

The survey results illustrated above indicate the general balance between the
policy alternatives of further mobilization and demobilization. Peak support for
escalation amongst domestic political actors occurs roughly at the time of
Westrnoreland’s favorable statements regarding the likelihood and proximity of obtaining
the issue-goods benefits. Similarly the level of support for bombing campaign against the
DRV had declined, moving from 63 percent in December 1967 to 24 percent by April
1968 (Mueller 1971, 370).140 The share of the public advocating greater mobilization
declines markedly over the next year, but still outweighed support for withdraw. A
survey in 1969 found that while 80 percent thought the war a mistake only 36 percent
favored immediate withdraw (Lunch and Sperlich 1979, 24).

Tet lowered the utility of the war by reducing perceptions of the likelihood of
achieving the issue-goods benefits of fighting as well as increasing the likely time until
these benefits would be realized. However the reaction after this shock to appraised
utility is different conditioned upon how different domestic political actors weight costs
and benefits. The hawkish cohort of the public believed greater effort not less was
needed in Vietnam as well as a growing chorus of doves wanting to withdraw.

While Hanoi’s General Offensive served as a catalyst for reappraisal this is only
the culmination of political pressure building up before January 1968. The costs of the

war had eroded support for the executive and his war policies before Tet (Jacobs and

 

"0 The Rolling Thunder punishment campaign did not break Hanoi’s resolve or effectively interdict flows
of troops and supplies (Clodfelter 2006, 88-92).

147

Shapiro 1999, 592). By the beginning of 1968 over 16,000 Americans had died in
combat. Defense spending consumed nine percent of GDP and of this roughly three
percent was attributable to the war in Vietnam (Oberdorfer I971, 81). As Vietnam
became more unpopular generally this engendered increasingly pessimistic opinions
regarding the economy (Northrop 1996, 789). The costs of mobilization grew ever
greater, political pressure built articulating itself in both calls for further mobilization as
well as demobilization.

To use a crude simplification, doves were displeased by the costs of the war as
well as the seeming lack of progress in concluding a negotiated settlement.141
Conversely, hawks sought to achieve victory relatively quickly. While frustrated at the
costs of the war the latter group appeared to be dissatisfied with relatively incremental
mobilization and seeming unwillingness to defeat the communist forces.

By failing to substantially alter war policies, other than incremental increase in
mobilization, Johnson did not successfully mitigate political pressure. Facing the reticent
North Vietnamese no substantial movement toward a negotiated settlement was made by
Hanoi.I42 Johnson also showed little interest in relinquishing the key issue-goods
objectives being pursued. Without guarantees of Saigon’s autonomy the American

executive was not especially interested in negotiations. Perhaps most vividly illustrated

by his willingness to have unconditional “discussions” but without entertaining serious

 

”l A point illustrated by the disappointed doves staying home during the 1968 election (Schreiber 1973).
"’2 “Thus, while Mr. Johnson realized that negotiation was increasingly a domestic political necessity, he
did not think it would lead to a settlement of the war. The president’s skepticism had such an effect on the
search for negotiations that, as one aide put it, ‘Every time we entered into a bombing halt, every time some
third party reported to us that there might be a chance for talks with Hanoi, and every time Hanoi
encourage American citizens visiting North Vietnam to believe that it was willing to enter into productive
talks, the president would just figure the odds. He never once saw a moment when, if he had been Ho Chi
Minh, he would have responded positively to our effort to start negotiations. But we had to have some
bombing pauses to show our critics at home and abroad that we were willing to take risks for peace and it
was Hanoi, not us, that was putting up obstacles.” (Goodman 1978, 28).

148

compromises of American issue-good positions (Kaiser 2004). The occasional bombing
halt designed to coax North Vietnam into negotiations were insufficient to garner a firm
commitment to meet. Consequently besides the rare secret meetings authorized by
Johnson no negotiations were undertaken.

The executive also demonstrated an unwillingness to seriously change strategy.
The Rolling Thunder campaign was clearly not affecting the flow of men and supplies to
the South nor undermining the DRV’s morale. Yet the executive feared broadening the
war and would not deviate from this relatively mild punishment strategy (Clodfelter
2006, 141-142). The attrition strategy of attempting to destroy more communist soldiers
than could be replaced resulted in heavy losses for NVA and VC soldiers. But American
casualties were also high, especially for a war against a tertiary communist power. Even
when pervasive criticism was leveled against the commander Johnson maintained support
for Westrnoreland (Sorley 1999, 9). Moreover the recommended substitution of US
combat units with ARVN was not contemplated until June 1968, after the 31 March
announcement that the executive would not seek reelection (Sorley 1999, 38-39).

Domestic political actors began to assess the war’s net utility comparatively
unfavorably but this was not effectively addressed by the American executive. Neither
change in negotiation position, change in military strategy, or rapid alteration in level of
mobilized resources occurred. Consequently Johnson’s support collapsed as efforts to
improve the utility calculation of his principals was not undertaken previous to Hanoi’s
Tet Offensive.

Richard Nixon defeated a fi'actured Democratic Party by offering demobilization

of the war effort in Vietnam, while still garnering the principal issue-good benefits in the

149

form of an independent South Vietnam, a policy plank known as “peace with honor”
(Kimball 2004, 24). In the aftermath of the Tet Offensive the Viet Cong had lost much of
its political infrastructure. The new MACV commander Creighton Abrams attempted to
build upon these successes, by using allied troops to control territory and deny recruits
and resources to the insurgency. Objectively the likelihood of achieving a viable pro-US
regime in the South was more plausible than ever (Kimball 1998, 259). A point
recognized by the president, but the executive’s assessment of the benefits of maintaining
an independent RVN appeared to be substantially higher than most of the public.143

Even with an improved strategy for pacification of the Vietnamese countryside,
Nixon understood that achieving his preferred issue-goods benefits would be difficult.
Talks had been started after Johnson’s announced bombing halt, but these had quickly
gone nowhere (Goodman 1978). It was clear to the new president that Hanoi was more
than willing to let the clock expire on the American public’s patience rather than offer
meaningful concessions in a negotiated peace.

US public opinion was bimodal in its disposition toward how the war should be
prosecuted, offering a dilemma for the new president. How best to accommodate
domestic political actors more favorably disposed to the war’s utility while not alienating

the somewhat larger cohort of doves demanding progressive demobilization.144 In order

 

”3 “The US government had been and continued to be committed to the preservation of a noncommunist,
pro-capitalist government in an independent state of South Vietnam, whether it was truly democratic or not.
Communist and allied Vietnamese nationalists were committed to the expulsion of US forces and the
reunification of North and South. Thieu [RVN President] sought to win agreement on the unconditional,
unilateral withdrawal of US troops and the removal of Thieu from power—although, as events proved, the
latter goal was not quite as important to the DRV and PRG as unilateral American withdrawal.” (Kimball
2004, 28).

“Despite talk of having neutralized the Vietnam issue, Nixon was becoming concerned about the long-
term impact of the war on the solidity of his political base as he looked forward to the 1972 presidential
election. Kissinger claimed that the president believed the war “was sapping his domestic support and
therefore had to be ended before 1972. . . Nixon’s political base included those who had been willing to

150

to mitigate political pressure the costs of the war were reduced, largely this was achieved
through the policy of Vietnarnization.

Under this program US troops were to be replaced by a better trained and
equipped ARVN. As the number of American troops and casualties declined it was
believed that the public’s assessment of the war’s costs would also decrease, permitting a
longer commitment to the RVN. By the end of 1969 the US troop presence had
decreased by 68,000 to 484,000 personnel, but battle deaths remained high at 9,414
(Clodfelter 1995, 171).

After articulating the American bargaining position in May 1969, which included
the total withdraw of US soldiers, the executive found that Hanoi was in no mood for
negotiations and regular meetings in Paris were suspended for the lack of appreciable
movement (Levy 1995, 154-156). While the Paris talks managed to give some indication
as to the administration’s willingness to conclude the war, lack of progress disappointed
dovish domestic political actors. Secret talks between Kissinger and North Vietnamese
fared little better (Kimball 1998, 103).

As the glacial pace of negotiations reflected a deeply resolved opponent Nixon’s

frustrations were articulated in a November 1969 speech.145 The 1972 presidential

 

trust his promises that de-Americanization, Vietnarnization, and negotiation would bring about a timely
peace with honor, but it also included those hawks who believed that the application of grater military
force, particularly more bombing, could achieve the desired end.” (Kimball 1998, 229-230).

I""‘Internal PRG [Provincial Revolutionary Government the communist shadow government in the South]
planning papers and assessments later indicated that from Hanoi’s viewpoint far too little progress had been
achieved in these tasks to create the position of strength essential for substantive discussions with
Washington. A related problem was the limited extent of the PRG’s population control in the south—even
the most conservative US estimates put the figure at less than 25 percent. These concerns by no means
preoccupied the entire North Vietnamese Politburo, for some members presumably believed that Hanoi
should never negotiate and, therefore, the question of achieving a position of strength for such a purpose
was irrelevant. Members of the Politburo who subscribed to this view probably argued simply that what
had not been won on the battlefield could not be won at the conference table. Other members of the
Politburo presumably favored negotiations, but only insofar as a negotiated settlement provided the United

151

election loomed large and time was not on Nixon’s side, even if a de jure settlement was
not imminent the executive devised a change in strategy that was meant to improve the
RVN’s prospects for survival.

Vietnarnization continued to reduce US fatalities throughout 1970, but also
reduced American combat power. In order to heighten the likelihood of securing a viable
Saigon regime, an incursion into Cambodia was devised. This was coupled with
intensified bombing of the Ho Chi Minh Trail that was principally located in “neu ”
Laos and Cambodia (Randolf 2007, 161-162).

The invasion of Cambodia in April 1970 was designed to destroy the NVA’s
central command center for operations in South Vietnam. Spearheaded by some of the
remaining US combat forces, substantial numbers of communist troops were not engaged
although large caches of supplies and munitions were discovered. While the NVA’s
command center remained elusive, the operation claimed to have degraded communist
military capabilities in the South (Kissinger 1994, 693-694; Clodfelter 1995, 176-184).

However the invasion of Cambodia also reinvigorated the flagging anti-war
movement. The unpopularity of the war led to changes in policy positions or defeat for
more hawkish members of congress (Gartner, Segura, and Barratt 2005/2006). Despite
reductions in the number of American troops in theatre and casualties, the midterm
elections resulted in an increased number of anti-war votes in Congress. As polls
indicated that for the first time that dissenters of war policies now favored withdraw more

than escalation (Lunch and Sperlich 1979, 28).

 

states with no more than a face-saving way to withdraw its troops and to end its support to the Thieu [RVN
President] government.” Kimball 1998, 104

152

A forced demobilization by the legislative branch was a genuine possibility even

as costs declined.I46

The American executive looked to maintain the support of those
more hawkish domestic political actors and achieve his preferred issue-goods position of
an independent RVN. These diminishing, but still significant, hawks were a key
constituency in his 1968 election victory and would also figure prominently in his
reelection bid (Schreiber 1973, 93).

Understanding the threat of deselection as well as a repudiation of his preferred
war policies through demobilization, Nixon ordered a second assault into a neutral
country. This time the incursion into Laos was conducted exclusively with South
Vietnamese troops. As with Cambodia, the assault designated Lam Son 719, was
designed to degrade the logistical capabilities of the communist forces in the RVN.
Additionally the operation would also attempt to disrupt preparations for the next major
NVA offensive anticipated for the dry season of 1972 or 1973 (Kimball 2004, 28-29).
The February 1971 attack made initial progress but stalled under incompetent leadership.
An NVA counterattack routed the ARVN which was forced back into South Vietnam.

This embarrassment for Nixon’s Vietnarnization program did little to reassure the
American public of the probability of achieving the issue-goods of a viable Saigon
regime even as costs declined political pressure mounted. ’47 The dovish portion of the

American electorate garnered limited benefits from further fighting, as the likelihood of

achieving a viable RVN seemed reduced.

 

"6 “In April and May [1971] demonstrators mounted impressive rallies against the war in Washington and
across the nation; in June the New York Times published portions of the Pentagon Papers; and the growing
strength of antiwar dissenters in Congress caused the White House to take seriously the possibility of a
cutoff of funding for the war or legislation calling for its termination ‘at almost any time’.” (Kimball 1998,
249)

”7 American combat deaths were only 1,381 for 1971 (Clodfelter 1995, 258).

153

Further the utility of the issue-goods at stake also seemed to be less esteemed as
the Saigon regime was revealed to be corrupt and autocratic. Also revelations of
American atrocities, most notably at My Lai, raised doubts regarding the justness of the
war (Young 1991; Wells 1994). In fact even as American troop levels were reduced the
level of support for mobilization continued to diminish (Levy 1995, 161).148 Nixon
understood that he needed to secure as much issue-goods benefits as possible before
Congress refused to mobilize or his reelection campaign was jeopardized.

Revising his issue-goods position the American executive acquiesced to a major
concession by permitting the presence of NVA troops in the RVN after the US withdraw.
Substantially this meant a shift in objectives from maintaining a US ally in the RVN to a
substantial lag from the final American military withdraws until the collapse of the
Saigon regime. Endogenous to Nixon’s calculations of the benefits of prosecuting the
war, was to illustrate the validity of an American alliance and demonstrate resolve. Even
if Saigon fell, the greater part of the war’s utility might be realized by reassuring the US’s
Cold War partners as to the robustness of Washington’s security commitments.149

The American executive’s proposal did not elicit the hoped for response and
instead a counter-proposal that was patently unacceptable was made (Kimball 2004, 28-
29). Change in issue-goods position had not substantially altered perceptions of the war’s

utility for domestic political actors. As political pressure continued to grow another

avenue was sought by approaching the DRV’s benefactors.

 

”3 “The percentage of Americans who thought withdraws were happening to slowly moved from 26
percent in late 1969 to 34 percent in 1970, to 45 percent in May 1971, to 53 percent by November of that
year.” (Levy 1995, I61).

”9 “The goal, as previously, was to save Thieu [RVN President], but Nixon and Kissinger now accepted the
alternative of a decent interval after an American withdrawal. The South Vietnamese would then ‘protect
themselves.’. . .If they failed to obtain the desired agreement, they would trade the cessation of bombing
and the retraction of their commitment to US residual forces in a post-cease-fire South Vietnam for the
release of POWs.” (Kimball I998, 260).

154

Already pursuing a more constructive relationship with the Soviet Union and
China, incentives could be offered to leverage a more malleable attitude out of their

client. 1 5°

The efforts to pursue a thaw with the communist powers culminated in
preparations for trips to China and the Soviet Union, February and May 1972
respectively, in which Nixon would press his case for a negotiated solution to the
Vietnam War. However should the diplomatic initiative and issue-goods position fail to
reach a settlement the American executive intended to alter military strategy.

The president had long believed, at least since the campaign of 1968, that a
demonstration of determination in interaction with brute force would coerce the requisite
concessions from Hanoi for his honorable peace. The doctrine became known as the
“madman” strategy. Built upon the personalities of Nixon’s bad cop to Secretary of State
Henry Kissinger’s good cop, the latter would warn of the impending destruction that
North Vietnam invited by not coming to terms with the former (Kimball 1998,76).151

The latent hostility reflected in Nixon’s strategy was soon to be given an outlet through

 

"0 “Positive and negative incentives had encouraged each side to offer new proposals. Despite the setback
they suffered in the Laotian invasion, Nixon and Kissinger were hopeful that recent progress in talks with
Beijing and Moscow boded well for their strategy of using détente with the Soviets and rapprochement
with the Chinese to extract concessions from Hanoi in the Paris talks. If an armistice could be negotiated
with Hanoi before the end of the year 1971, they calculated, it would prevent the North Vietnamese from
launching an expected offensive during the 1972 spring dry season. . .On the negative side, there was
mounting pressure on the White House from congressional liberals to withdraw troops by a definite date
and growing impatience among congressional conservatives with the administration’s inability to conclude
the war.” (Kimball 2003, 28-29).

'5 ’ “By September 1971, Kissinger’s secret negotiations with Le Due Tho and Xuan Thuy in Paris had
reached a critical impasse, and on the domestic front congressional opponents of Nixon’s Vietnam policies
were stepping up their criticisms. Some complained about the administration’s involvement in the rigged
South Vietnamese presidential elections scheduled for October. Others called for the withdrawal of
American troops within six months. In an Oval Office conversation on September 17, three days after
Kissinger’s fruitless meeting with Thuy [DRV’s envoy] in Paris, Nixon angry with his critics and frustrated
with the war, talked about striking out at both his domestic and Vietnamese opponents. He would accuse
domestic critics of sabotaging his efforts to end the war on his terms, and he would deliver a ‘shot’ or a
‘pop’—that is, a brief but hard air strike—against the North Vietnamese in the DMZ. The purpose of the
latter was to ‘screw’ up, or ratchet up the war in order to stimulate a rally-‘round-the-flag response at home
while at the same time signaling Hanoi that he might do a ‘hell of a lot more’ in the future—a clear
statement of the madman theory.” (Kimball 2003, 193).

155

an overt attack by the NVA upon South Vietnam, which became known as the Easter
Offensive.

The Politburo’s decision to launch assault upon the RVN offered a critical test of
Vietnarnization. Equipped with Soviet tanks and heavy artillery, on 30 March 1972 the
NVA stormed through the demilitarized zone and reduced ARVN firebases on the border.
The South Vietnam looked to be in the midst of disintegration as army units fled south
and panic appeared epidemic. Nixon feared that his attempts to salvage “honor” from the
conflict was being crushed under the weight of the NVA assault. Determined to use
every option at his disposal, the executive would not be dissuaded from using his
“madman” strategy even by his upcoming trip to the Soviet Union.152

A massive armada of American airpower was reintroduced into Southeast Asia.
Two different missions were crafted one was to destroy the NVA invasion force with
tactical airstrikes. The second was a massive strategic bombing campaign targeting the
DRV designated Operation Linebacker. To execute the destruction of the North’s
infrastructure and ability to resupply its invasion force fighter-bombers equipped with
new avionics, electronic countermeasures, and guided munitions thwarted the DRV’s
once lethal air defenses (Randolf 2007). Nixon also demanded that as many B-525 as

could possibly be mustered should be used to target North Vietnam, with the intention of

horrifying the Politburo as to the consequences of further procrastination.

 

'52 “Hanoi thought the events scheduled for 1972 would force Nixon to use restraint in dealing with the
Easter Offensive this was not the case. “The North Vietnamese Politburo’s calculus that the summits and
the presidential election would restrain Nixon’s response to the invasion was the exactly opposite to
Nixon’s thought patters. The Moscow summit was, in fact, one of the stimulants to the decision to take
tough action; the timing of the convetions and the electoral campaign imposed urgency on aggressive
action that might not otherwise have existed.” (Randolf 2007, 165).

156

Tremendous damage was wrought on the NVA formations. The invasion of the
South was undertaken with large units fighting a pitched battle against the ARVN and
these conventional formations made for comparatively easy targets. The strategic portion
of Linebacker was also successful, appearing to be more effective than had been the
considerably longer but less intense Rolling Thunder campaign (Clodfelter 2006).
Nixon’s harsh tactics were also received with comparatively favorable reception by at
least some of the public (Kimball 1998, 316).153

American advisers also organized the ARVN defense and demanded that the
RVN’s executive President Thieu sack incompetent officers. These efforts saved South
Vietnam and after the initial NVA assault the ARVN accounted itself fairly well,
launching a series of counterattacks in the weeks that followed the invasion and retaking
all lost RVN territory (Truong 1980, 48-75).

Beijing and Moscow’s tacit endorsement of a negotiated settlement vindicated
Nixon’s efforts at détente.’54 While not necessarily crippling, the US efforts at
diplomatically isolating North Vietnam had been somewhat successful. This was
especially relevant given that the NVA was now more dependent upon foreign imports of
munitions and weapons, both of which were drawn down during the Easter Offensive.
These difficulties in conjunction with extensive mining of North Vietnam’s waterways

served to drastically curtail major military operations in the near future. '55

 

'53 Likely this was a rally effect associated with heightened military activity and supported by the more
hawkish portion of the public.

’5‘ “North Vietnam’s allies publicly condemned the bombing, but behind the scenes they advised Hanoi to
settle. China apparently reinforced this advice by threatening to ‘obstruct’ the delivery of additional, future
supplies to the DRV, but the Soviets assured Hanoi that it would pressure the United States to cease
bombing and return to the table as soon as possible.” (Kimball 2003, 366).

”5 The mining effort was called operation Pocket Money and was used concurrently with the heavy
bombing of North Vietnam (Randolf 2007).

157

Despite these successes, Congress attempted to legislate a rapid withdraw of US
forces ahead of the fall elections (Ambrose 1989, 299). Political pressure was still
mounting and again Nixon tried to leverage his preferred issue-goods out of negotiations
but at the same time cognizant of the prospect for deselection. Having bolstered support
for his war policies amongst the hawkish public.

The ongoing secret negotiations between Kissinger and the North Vietnamese
appeared to have achieved a breakthrough. Hanoi dropped its long-standing demands
that the US overthrow Thieu on their way out of the country and that a joint government
be formed that included the VC’s political wing the National Liberation Front (N LF ).
However, the NLF was to become the effective state authority in the territory it
controlled in October 1972 (Kimball 1998, 32). The details of the coalition government
and the role of US involvement withdraw remained tricky points of contention, but the
American delegation believed they had a deal.

The steady reduction in costs and seemingly imminent peace helped mollify
dovish criticisms of Nixon’s war policy. His determination to garner some issue-goods
benefits helped ameliorate more hawkish dissent and the American executive won the
1972 election overwhelmingly.

Before inauguration day, the peace talks were scuttled by Thieu who found the
terms patently unacceptable and demanded major revisions of the settlement. This
initiated a new set of demands from Hanoi and in effect caused the collapse of
negotiations. Nixon was pressed for time, mainly by the opening of the new
Congressional session in January and with it the long feared cut in war appropriations

(Lipsman et al. 1985, 12).

158

Nixon restated his deadline for negotiations set for the December. Hanoi was not
to be persuaded and the American executive launched Linebacker II in an effort to coerce

his antagonists into returning to the table.156

These attacks were widely condemned and
caused massive destruction. But the punishment strategy had the intended effect as the
traditionally adamantine resolve of the Politburo appeared to be unnerved. The DRV’s
air defenses had already degraded during the Easter Offensive and the surface to air
missiles capable of hitting B-523 were nearly expended in the first few days of
Linebacker Il (Clodfelter 2006, 179-202). Soon Hanoi was defenseless and
accommodated American demands for a peace.

In order to maintain public support for engagement in Vietnam casualties were
reduced through a withdraw in US troops, at the expense of the likely issue-goods that the
executive would have preferred. Alternative strategies included massive bombing of his
opponents undertaken to coerce Hanoi into compromise, even at the potential expense of
the executive’s efforts for détente helped bolster support amongst the more hawkish
cohort of domestic political actors. Nixon did not obtain his preferred issue-goods
position instead he accommodated dovish sentiment by incrementally reducing his offers
in order to facilitate a negotiated settlement. However by bowing to domestic political

pressure the American executive was able to avoid Johnson’s fate or a congressionally

mandated demobilization.

 

"6 “According to the US Air Force, the objectives of the two operations differed substantially. Linebacker
I had mainly been carried out as an interdiction campaign directed at the DRV’s supply system.
Linebacker II was aimed at applying ‘maximum pressure through destruction of major target complexes in
the vicinity of Hanoi and Haiphong’ in order to inflict sever damage to the DRV’s ‘logistic and war
supporting capability’ and make a ‘psychological impact’ on North Vietnamese ‘morale.’ Only 12 percent
of the sorties were against strictly military targets—airfields and surface-to-air missile sites.” (Kimball
2003, 279).

159

Vietnam is often viewed as the embodiment of how a democracy enjoying
superior capabilities is beholden to an unresolved public which forestalls victory in a
protracted war. Like the colossus with feet of clay, casualties enervate the willingness of
the public to mobilize. However while the public did appear to retract support for the
war, for much of the conflict critiques of the executive’s war policy favored escalation.
Demands for demobilization did grow throughout the war, but the percentage of wins that
the United States and South Vietnamese military enjoyed is actually higher at the end of
the war than at the beginning.’57

Another complication for the received literature is that if a post-hoe review of war
is thought to end in electoral sanction conditioning executive behavior, the watershed
moment for the public was the Tet Offensive. While these events helped foster the
removal of Johnson, this series of battles occurred nearly a five years before the
conclusion of the war. Accepting that an intra-war assessment is operable the public’s
critique of war selection takes place as the conflict becomes more difficult or longer than
anticipated. Selection of a war is therefore important in determining the level of public
support, but this does not explain how costs are offset by the benefits in the calculation of
domestic political actors.

By altering the cost-benefit calculations of the public through changes in
negotiation position and mobilized resources support for involvement in Vietnam, the
war was permitted to continue without the electoral sanction of Nixon. Similar to the
finding in the autocratic cases the constituents of the executive could be placated through

changes in war policy. The alteration of war policies rather than war selection is a more

 

'57 The quantitative chapters define battles and wins, but in the data coded for my dissertation the winning
percentage of the United States in Vietnam does not move monotonically downward.

160

cogent explanation for the continued US mobilization in Vietnam even following the Tet

Offensive.

United States: Korean War

The Korean War began in June 1950 as an attempt by the Democratic Republic of
Korea (DRK) to force reunification upon the US affiliated Republic of Korea (ROK).
While initially popular with the American public, heavy casualties and economic
expenses, contributed to a precipitous decline in the war’s utility. Furthermore public
opinion became divided as to how best prosecute the war either with greater mobilization
or by withdrawing. The failure to address either one of the two major critiques of the
executive’s war policies eventually led to Truman’s deselection.

Interpreting North Korea’s attack as one of Stalin’s machinations, President
Truman dispatched an army expeditionary force from Japan against the advancing
Soviet-equipped army. Despite the presence of US forces the North Korean forces
herded the defenders into a perimeter around the South Korean port city of Pusan. Aided
by airstrikes on North Korean logistics and further reinforcements, the defenses
surprisingly held.

With the Soviets boycotting the Security Council an endorsement of the US
intervention was obtained. The commander of the newly-formed multinational force was
General Douglas MacArthur. Having stabilized the UN forces’ position around Pusan,
on 15 September 1950 MacArthur orchestrated an aggressive attack with a secondary
force. The surprise amphibious assault landed at Inchon, near Seoul. The rapid advance

of the UN forces fi'om the beachhead severed the North Korean army from its source of

161

supply. The communist forces already exhausted and depleted by losses in the Pusan
action, were routed back northward. With the defense of the young ROK assured, a new
set of issue-goods where pursued that sought a unified peninsula under the rule of Seoul.

The pursuit of this new objective was initiated by MacArthur, but tacitly
acquiesced to by Truman. The prospect of an American ally fashioned in part from a
Soviet client considerably bolstered the net utility of the war to many domestic political
actors (Steuck 2002, 98). The optimism appeared to be well placed as October witnessed
the fall of the DRK capitol Pyongyang and organized North Korean resistance began to
crumble. Heedless the UN commander sought to finish the conquest quickly by
aggressively pressing his forces toward the Chinese border.

Given the nearly ubiquitous rally effect, unsurprisingly the American public was
favorably disposed to the war at the onset. All the more so when it appeared that the war
would be short and that communist domination of the DRK would be reversed. A survey
taken in July 1950 found that only 14 percent of respondents believed that the war would
go on longer than a year (Mueller 1971, 361). Both the likelihood of achieving the issue-
goods in question and the time until these benefits would be realized made for relatively
favorable impressions of the war’s utility. The realization of reunification would become
much more difficult than was apparent in the fall of 1950.

As UN forces began to press upon the Yalu River, amassed at the border was a
force of 260,000 Chinese (Korean Military Institute 2001b, 109-115). While intelligence
sources had reported their presence these had been dismissed by the MacArthur.
Beijing’s euphemistically titled People’s Volunteer Army (PVA) was composed of

veterans from the recently concluded Chinese Civil War. Aided by surprise and by the

162

poor disposition of UN troops, the Chinese Communists executed their practiced
maneuver strategy.

In late November 1950 this force struck with unanticipated strength and skill. An
enormous press of PVA soldiers attacking the UN positions was combined with
infiltration behind the front. Smaller units were sundered from their parent formations
and the intense logistical needs of the American forces were threatened by the presence
of Chinese troops behind them (Li, Millet, and Yu 2001). This initial PVA assault was to
be the first of five “spring offensives”.

Soon after the UN forces were reorganized and conducted a hasty but ordered
retreat, using smaller units to screen the bulk of MacArthur’s southbound forces.
American airpower savaged the pursuing PVA and the superior mobility of the American
equipped army permitted a fighting retreat ending south of Seoul. The third Chinese
spring offensive resulted in the loss of the southern capitol in January 1951 in
conjunction with earlier defeats at the hands of the Chinese Communists catalyzed a
reappraisal of the war (Feaver and Gelpi 2004, 137-140).

Even before the onset of the Korean War the Truman presidency was already
tarnished. The Communist victory in China and the Soviet development of the atomic
bomb in 1949 both diminished the executive’s political standing (Theoharis 1970, 196-
219; Caridi 1968). Evidence of espionage in the federal government, accurate or not, fed
paranoia of communism and the demagoguery championed by Senator McCarthy. More
specifically a substantial minority of representatives and senators were staunch advocates

of the Nationalist Chinese cause and Generalissimo J iang J ie-shi (Koen 1974).158 This

 

'5’ Frequently referenced by the pejorative term the “China Lobby”.
l 63

sentiment was also shared by a substantial portion of the broader public that demanded a

strong response to what was perceived as Red China’s aggression.

Figure 10: Lowess Graph of Public Support for Korean War and War Duration159

 

 

70 a,

|
Aug-50
60 ;
Percent
Approval .
of Korean l
War

50 a

l
_ Aug-5|

l Apr-51
f Dec-50 """"" ~ ..

4o .1 Feb-51 Mar-51 Jun-5| __
~ 1‘ Sep—52
1 Mar-52 Oct-52

30 .1
o lo 211 36

War Duration (Months)

bandwidth = .8

As was the case with the Vietnam, public opinion of the Korean War was
bimodally distributed with both a more hawkish and dovish cohort of domestic political
actors. But the political milieu of the early-19505 appears to have contributed to a
preference for more hard-line policies at the beginning of the war (Pierpaoli 1999, 110).
A survey taken in December 1950 asked respondents what policy they advocated
following the entry of the Chinese into the war: 28 percent supported withdraw, 25

percent supported intensification of the war effort, and 27 percent advocated the use of

 

"9 Data come from Mueller (1973) reprinted in Feavar and Gelpi (2004, 137).
l 64

the atomic bomb (Mueller 1971, 367). Whereas the American public was dissatisfied
executive policy they were conflicted as to what policy options to pursue.

The UN forces required enormous supplies that encumbered them during their
southward flight. However these units also enjoyed a massive advantage in firepower.
The PVA’s fifth offensive, resulted in massive casualties for communist soldiers. After
having stopped the Chinese in the spring of 1951, the American-led forces took the
strategic initiative, recapturing Seoul in March and pressing the PVA north. Even with
momentum on their side and enjoying a great deal of success against their weakened
opponent, in May the offensive was halted by Truman.

The reduced likelihood of an expanded war and freeing up resources to be
employed elsewhere appeared to influence the executive’s acceptance of an ante bellum
status quo resolution (Halperin 1963, 20). The increasing ntunber of casualties was also a
consideration. By the end of April 1951 the US had already suffered over 18,000 battle
deaths.I60 Even if the UN forces could achieve a unified Korea as an American ally, the
costs and the likely time until this could be realized were raised substantially by Beijing’s
intervention. The executive’s shift to an ante bellum status quo issue-goods position
meant that a change in strategy could be accomplished. An attrition strategy was
designed by creating static fortified positions, reducing US casualties while making
communist assaults as costly as possible (Korea Military Institute 2000b, 715).

The decision corresponded to a plurality of domestic political actors preferences.

Asked in a March 1951 survey, 43 percent approved of a status quo settlement and 36

 

"’0 Mueller finds that after the initial bout of high casualties that the marginal increase in battle deaths does
not appear to alter the level of support for the war (I971, 366).

165

percent opposed if the communists agreed to stop fighting (Strunk 1951, 390)."5| The
decision to seek a status quo peace and to reduce battle deaths seemed to be roughly
attenuated to the plurality of more dovish domestic political actors. However another
large cohort supported greater mobilization for increased issue-goods benefits and were
given voice through elite criticism of the executive.

Douglas MacArthur was one of the most prominent supporters of pursuing victory
through a unified Korea. ’62 Serving as the UN commander did not preclude MacArthur’s
lobbying for the pursuit of these more salient issue-goods. Having already once initiated
an effort to destroy the DRK, the General suggested that a naval blockade of China,
strategic bombing of logistical points in Manchuria, and potentially the introduction of
Nationalists troops from Taiwan should be considered (Caridi 1968, 432). These
sentiments were quickly made public and were justifiably viewed by Truman as gross
insubordination. After a series of clashes with the executive MacArthur was relieved of
command in April 1951, but the popular figure did not go quietly.""’

The advocacy of alternatives to Truman’s war policies appeared to resonate with
those more hawkish domestic political actors. Echoing MacArthur, Southern Democrats
and Republican members of Congress began to assert that a limited war would not enjoy

the support of the public and instead victory with its greater issue-goods utility should be

 

“’1 A March opinion poll indicated that 67 percent of the public supported staying in Korea whereas 20
percent favored withdraw (Casey 2008, 215). The Chinese intervention had made the likely time until a
settlement would occur almost certainly longer and was found to negatively reduce the level of support for
the duration of the war (Mueller I971, 361).

'62 MacArthur was a Republican and viewed as one of the party’s view national figures after losing five
consecutive bids for the White House.

"’3 A modification of the present containment policy became a staple of partisan politics throughout 1951.
suggests that this debate was first initiated by former president Hoover’s assertion that a limited number of
strategic allies should be maintained. That the main Eurasian land mass be allowed to fend for itself and
strategic positions like Britain and the Philippines be maintained as bulwarks against the USSR Extremes
aside other policy makers quailed at the drain that US resources were imposing on themselves to ensure
that in a limited conflict and pitted against a proxy foe (Piperpaoli 1999, 56-57).

166

pursued (Caridi 1968, 426-429).“54 This policy critique reached its zenith when
MacArthur’s firing became the subject of a senatorial hearing.

Giving traction to elite criticism of executive war policy was the exploding costs
of the war. Heavy casualties were undermining popular support for the war (Mueller
1971, 361). Military expenditures undertaken to fight the Korean War, along with other
Cold War commitments, massively increased defense spending. In 1949-50 the US
defense budget was $13 billion in two years it quadrupled to $52 billion (Pierpaoli 1999,
30). Government expenditure on national security would reach 10 percent of GDP, this
increase in military spending fueled inflation. To combat depreciating purchasing power
wage and price freezes were implemented. Despite these efforts inflation persisted
throughout 1951 and the regulations became a major irritant to both business and labor
interests (Pierpaoli 1999). “’5

The increase in costs and the reduction of issue-goods benefits helped precipitate
political pressure being applied to the executive (Stueck 2002, 157). The elites and the
hawkish public embraced a policy critique that the sacrifice in lives and monetary
expenditures did not justify the modest benefits of pursuing a status quo settlement in a
peripheral war. Political pressure coalesced around support for alternative policies,
including an expansion of the war to challenge the new communist regime in China and

thereby secure a unified Korea and perhaps destabilize the Beijing regime.

 

164After the MacArthur hearings, a Republican minority report declared: “We believe that a policy of
victory must be announced to the American people in order to restore unity and confidence. It is too much
to expect that our people will accept a limited war. Our policy must be to win. Our strategy must be
devised to bring about decisive victory.” (Halperin 1963, 24-25).

“’5 Organized labor was also antagonized by Truman’s executive order to compel striking workers to return
to their jobs (Pierpaoli I999, 59).

167

As shown in the graph below, support for aggressive moves against the People’s
Republic of China appeared to enjoy popular support. Surveys from the time reveal
increasing willingness to accept an ante bellum status quo settlement, which increased
from 43 to 51 percent between March and July 1951 (Casey 2008, 263). The somewhat
contradictory responses may have reflected a portion of the public that saw greater
mobilization as a means of shortening the war, but without achieving the destruction of
North Korea. Although political pressure for more hawkish actions waned somewhat
after the MacArthur hearings, a substantial portion of the American public resented

“dying for a tie” (Stueck 2002, 157).

Figure 11: Percentage Supporting Bombing China and
Assisting Nationalist Invasion of Mainland

60 4 o Oct-5|
l
0(month) 58 11 Apr-51

Sep-S 1

Percentage
Support for
Bombing China I

(month) 1
Percentage56 '7 May-5t Mar-52
Support for
Assisting
Nationalist
Invasion of l M ar-52
China 54 Aug-51 °

—‘ C

 

 

L—Y— _ ___ ‘_ ______-

0 10 20 30 40
War Duration (Months)

168

The opening of negotiations offered a chance to pursue the executive’s preferred
policies. Armistice discussions begun July 1951 provided an opportunity to improve the
more dovish domestic political actors’ perceptions of the war’s utility, through the
realization issue-goods more quickly. However the peace talks did not progress well.

The Chinese delegation, which dominated the negotiations, requested that
peripheral issues be considered in an armistice. The issue-goods demanded included
granting the Chinese Security Council seat to the People’s Republic and concessions with
respect to the American security relationship with Taiwan (Xia 2006, 51). The UN
delegation had no interest in linking war termination with these other issue areas and
negotiations consequently stalled.

In December 1951 the Chinese dropped their demands for a resolution of issues
not directly related to the war. The basic framework for the division of the peninsula
roughly along the 38‘” parallel was arrived at fairly quickly (Zhu 2001, 77-79). A new
obstacle to a settlement was encountered on the question of the repatriation of prisoners
of war (POWs). The US refused to turn over communist prisoners who did not want to
return to North Korea or China, esteeming the propaganda value of communist soldiers
not wanting to return to a life under tyranny. Talks again ground to a halt.

Although successful in mitigating political pressure amongst the more dovish
portion of domestic political actors, as talks stalled discontent with war policies grew.
Even after settling into a defensive attrition strategy, the UN forces continued to accept
casualties prosecuting the war. These costs along with the enormous economic
expenditures for mobilization increased political pressure dramatically (Gartner and

Segura 1998).

169

The American executive continued to pursue the ante bellum status quo issue-
goods. Truman’s preference for an intermediary settlement had reasonably strong
support, but as negotiations stalled over a comparatively minor security issue this
detracted from the benefits of further fighting. This dissatisfied the more dovish
domestic political actors who saw limited benefits being obtained by holding out for
voluntary repatriation of POWs relative to the costs incurred. The more hawkish cohort
of domestic political actors resented fighting a limited war for modest objectives (Caridi

1968; Stueck 2002, 157).

Figure 12: Lowess Graph of War Duration and Monthly Battle Deaths166

 

 

 

4000 g
'l
l Nov-50
3000 l 569-50
J
' Jul-50
Monthly US
Battle Deaths
2000 n] Dec-50
Feb-5|
l Oct-51
Au -50
1000 g “~~~.,May-51 369'“
n Oct-52 Jul-53
l Apr-51
‘ Mar-51 -~ Jun-52 Sop-52 Mar-53
l Jan-5| Jun-51 Aug-52 _, _,__. -_---- Jun-53
I Oct-50 Au -51Nov-51Feb-52 ’ ' " ’ Nov-52 Feb-53A r-53
1 Jul- Jan-52 Apr-52Jul-52 Dec-52 ay-53
0 “‘__F_J_I_I_II:_50 - Dec;5_l_ Mar-52May-52 . Jan-53 _._
0 10 20 30 4O
bandwidth = .8 War Duratlon (Months)
166
(Clodfelter I992).

170

The American executive did not accommodate this burgeoning political pressure
with a significant movement from his preferred war policies. For moral or propaganda
purposes the repatriation of all communist POWs was not an issue that the executive
planned on compromising. The ossification of issue-goods position was mirrored in the
static front lines, reminiscent of the Western Front of World War I.

The decision to halt the offensive against the communists in May of 1951 helped
reduce UN casualties. However the attritional struggle that followed meant that both
sides dug into rugged terrain that made any future attacks by the American led forces
comparatively expensive in terms of expected casualties. Consequently it became
militarily more difficult to coerce a settlement from their antagonists (Stueck 2002, 157-
158; Foot 1990, 62-63).

In order to avoid an expansion of the conflict amphibious landings behind enemy
lines or attacking China’s war making ability were precluded. This made compelling the
communists to settle through military force more costly and difficult. Truman also
wanted to avoid tying down conventional military assets in a peripheral conflict (Halperin
1963, 23). But so long as the relatively resolved Chinese and North Koreans were
confident that they would retain nearly all of their issue-goods utility through a protracted
war, there was little incentive to compromise with their casualty averse opponent.167

Truman’s failure to alleviate the high levels of political pressure of either the

dovish cohort of domestic political actors or the sizeable more militant faction meant his

political collapse. The unwillingness to alter strategy to improve perceived benefits,

 

"’7 The Soviets encouraged the hardening of the Chinese negotiation position in the hopes of tying down
the United States in a peripheral conflict (Xia 2006, 55). The communist delegation’s rigidity was
encouraged by Moscow. The Soviets lobbied for a hard line stance with the UN delegation, in the belief
that the US was less tolerant of the costs of the war than was China (Stueck 2002, 157-158).

171

demobilize the war effort, or reduce issue-goods negotiation position resulted in his
deselection. Republican Dwight Eisenhower replaced Truman and planned to terminate
the war with alacrity.

Like his predecessor the new president was unwilling to acquiesce to the
communist demands that all POWs be repatriated. In order to achieve a quick
termination of the conflict a change in strategy was made. The grinding struggle of
attrition between the antagonists had continued for nearly two years, with little
perceptible movement toward an agreement. An effort to increase the costs of China-
North Korea’s intransigence was undertaken through an intensification of the air war,
something Truman had avoided in part out of fear of broadening the conflict (Clodfelter
2006,13)

American airpower had mainly been used in support of ground troops or
interdicting the flow of supplies, this changed in May 1953 when a sortie of fighters
destroyed an irrigation dam in North Korea. The operation caused a major flood that
destroyed a number of roads and bridges. More importantly it served as a warning of
American determination to expand its package of targets and launch a harsh punishment
strategy to end the war (Foot 1988-89, 110; Clodfelter 2006, 17).

The effort to coerce the Chinese into an agreement was not limited to strategic
airstrikes. The US began to posture as if it were preparing for an amphibious invasion
north of the 38‘h parallel. The invasion would be part of a broader escalation of the war
that might include a blockade of mainland China. Eisenhower also made thinly veiled
threats of nuclear strikes against rrrilitary targets in Manchuria if the Chinese did not

come to terms amenable to the United States (Foot 1988-89, 97—98). Even if the threats

172

were not entirely credible, the Chinese leadership was already war weary and an
expanded conflict did not appeal to Beijing. An armistice was concluded in July 1953
with the voluntary repatriation of POWs.

The Korean War offers an example of domestic political actors’ bimodal
preferences in terms of war policies. A June 1953 poll indicated that 45 percent of
respondents said that an armistice with China would be considered a success whereas 69
percent thought this was the best course (Edwards 2006, 170). A substantial cohort of the
public sympathized with the need for an expanded war for greater issue-goods benefits,
while another portion of the citizenry wanted a quicker settlement and a reduction in the
war’s costs. In the end Truman accommodated neither group, electing instead to pursue a
moderate issue-goods settlement in compliance with his risk aversion to a broader war.
While the executive did change tactics to a lower cost attrition strategy, this also
distended the conflict as military pressure was not being heavily applied to the
communists.

Eisenhower amended war policies to facilitate a termination of hostilities. New
strategies for punishing the intransigence of China and the DRK were threatened. The
potential for an expanded conflict in order to achieve a quicker resolution to the war
appealed to the more hawkish cohort of domestic political actors. The new executive was
able to achieve a negotiated settlement soon after.

As with Johnson in the Vietnam War, Truman was forced from office after having
joined an Asian conflict. In both cases the American president appeared to derive greater
utility from the issue-goods benefits than did his constituents. Having selected to fight

the war, the initiating presidents then demonstrated an unwilling to alter war policies for

173

the purposes of maintaining support for mobilization. This broadly conforms to the
depiction of a public that has a clock the patience for a war ticks away immutably toward
a refusal to further mobilize.

The immutable enervation of public resolve does not contextualize the offsetting
costs and benefits of the war, such that despite perpetuating the conflict for years Nixon
did not receive the same sanction as Johnson. As with the autocratic cases, domestic
political actors tend to ascribe less utility to an ongoing war than does the executive.
Evidence of success in the form of battlefield victories or enemy casualties can help to
demonstrate the likelihood of achieving a war’s objectives.

As the relative resolve and capabilities of the opponent are revealed through
fighting to both executive and the public, so is the utility of the issue-goods benefits and
likelihood of achieving these objectives. Casualties and war duration are not the only
factors in determining public support for a war and by the executive altering war policies
political pressure can be mitigated. These changes in war policy might then mute the
relative importance of war selection and instead make the conduct of the war the
mechanism by which a democratic executive’s accountability to the public is most

significant.

United States: World War 11

President Roosevelt had meticulously cultivated support for American
intervention but needed the attack on Pearl Harbor to obtain an endorsement for complete
mobilization. The objective of the war could be interpreted as an effort to prevent the

domination of two regions by hegemonic powers (Mearsheimer 2001). Accordingly the

174

implications for the future security of US meant that typically costs outweighed benefits
heavily in the minds of most domestic political actors.

The mandate that both Roosevelt and Truman received to prosecute World War
II, meant that the unitary assumption is more accurate in this case. However the way the
war was conducted was also done in anticipation of the potential for political pressure to
emerge, resulting in war policies that might have been somewhat different than the
executive’s unitary preferences.

The issue-goods at stake in the war were esteemed highly by American domestic
political actors, consequently discontent amongst the public for war policies was much
more muted than the peripheral Cold War conflicts. A February 1944 survey found that
77 percent of respondents did not believe that US involvement in World War II was a
mistake. A similar question asked in March 1968 and December 1950 found that only 41
and 39 percent of respondents felt the same way about Vietnam and Korea respectively
(Mueller 1971, 361; Lunch and Sperlich 1979, 23, 25).

By the time of the US entry into the war the Germans were nearly at the gates of
Moscow threatening to add the Soviet Union to their other European conquests.
Roosevelt appropriately designated Nazi Germany as the greater threat a move congruent
with public opinion at the time (Burns 1970, 179; Gallup and Fortune Polls I942, 486).
Hastily an American force landed in November 1942 into North Afiica hoping to divert
German units away fiom Eastern Front. ’68 After suffering minor setbacks the Anglo-
American forces soon dominated the German Afiika Korps and made preparations for a

1943 invasion of Southern Europe. More costly was the battle for the Atlantic in which

 

’68 The diversion of Wehrmacht resources to North Africa remained very modest, even after the July 1943
invasion of Sicily and Italy absorbed a small portion of the German military effort (Magenheimer 1999,
142)

175

the US Navy and merchant marine fleet struggled to combat the U-boat threat. As with
the campaign in North Africa the superior resources of the Western Allies eventually
prevailed, after serious casualties (Kennedy 1987, 348-356).

In the Pacific a string of embarrassing defeats on both land and sea had followed
the US disaster at Pearl Harbor. The Japanese had captured the American force on the
Philippines and conquered British colonial possessions, most notably Singapore. It was
not until June 4-7, 1942 when the US unexpectedly defeated a superior Japanese carrier
force at Midway that Tokyo’s expansion was checked. Without the benefit of a
substantial carrier strike force with which to take the strategic initiative the Imperial
Navy was put on the defensive.

The civilian and military leaders in the Roosevelt administration understood that
America’s superior industrial capacity could make good material losses, a luxury Tojo
did not enjoy.169 Operating with this logic the US began to implement an attritional
strategy of forcing naval battles and conquering islands progressively closer to the home
islands of Japan (Burns 1970, 305).

Defeats inflicted upon the Japanese navy and the Anglo-American invasion of
Italy helped to improve the prospects for victory. As the tide turned against the Axis
throughout 1943 the probability of achieving unconditional surrender appreciably
increased. By December 1943 over 90 percent of the public believed that the Allies
would win the war as compared to 72.5 percent in December 1941 (Cantril 1948, 39;

Gallup and Fortune Polls 1942, 152). Despite a salient set of issue-goods being pursued

 

“’9 Kennedy illustrates that Japan had only a tenth of the industrial capability of the United States on the
eve of World War II (1987, 330).

176

and a very high perceived probability of achieving these aims, fears as to the long-term

willingness to mobilize remained a concern for the executive.

Figure 13: Lowess Graph of War Duration and Approval of World War II

95 -1

F eb-43
90 J

Percentage

~ Sep-43
Public Febfz...

approval

Jun-45
85T Dec-41 Mar-43 ‘

Aug-44 ,/

so a May“

 

 

F eb-44

 

0 10 20 30 40
War Duration (Months)

bandwidth = .8

Some commentators suggested that public opinion while strong in the polls was
not of requisite intensity in order to fully prosecute the war. It was believed that this was
in large measure due to a lack of understanding as to what the objectives of the war were,
especially with respect to Germany which had not attacked the US (Stele 1978). In order
to continue support for mobilization a proclamation for unconditional surrender of the

Axis was issued at the January 1943 Casablanca conference. Roosevelt designed the

177

declaration to both clarify American objectives in the war as well as enhance the salience
of the issue-goods being pursued (Chase 1955, 274).

By making unconditional surrender the issue-goods position of the Allies the US
would have an opportunity to refashion the domestic politics of its antagonists potentially
mitigating the need for future conflicts (Frank 1999, 26-27). A theme Roosevelt revisited

in the Four Freedoms declarations.170

Improved public perceptions of the benefits of
fighting offset what was a very costly conflict.

At its peak, spending on the Second World War would consume 48 percent of US
GDP (Oberdorfer 1971, 81), and by September 1943 over 24,000 Americans had been
killed in battle. The Democratic Party took heavy losses in Congress during the 1942
midterm elections due in part to casualties and other costs imposed on the public (Cotton
1986, 628). But a survey taken in that sartre month revealed that the war was still viewed
favorably by 89 percent of respondents (Clodfelter 1992; Public Opinion Polls 1943).

Despite defeats meted out to his co-partisans Roosevelt remained popular,
enjoying 63 percent favorability rating in December 1943 and appeared to be in little
danger of deselection or popular failure to mobilize. Similarly a trial heat in the
September indicated that FDR led the eventual Republican nominee Thomas Dewey 55
to 45 percent (Public Opinion Poll 1943, 739).

Even with the high levels of support amongst domestic political actors, Roosevelt
implemented strategies that minimized the human costs of the conflict. Most notably a

massive bombing campaign was applied to Germany and its conquests. This punishment

strategy was a capital intensive operation but comparatively inexpensive in terms of

 

”0 The Four Freedoms represented a kind of proselytized liberalism that was to be transmitted throughout
the world in the wake of World War II’s conclusion. These freedoms included both political and economic
themes, potentially appealing to America’s sense of messianic purpose (Kissinger 1994, 33-41).

178

American casualties (Clodfelter 2006, 5-9).l7l While strategic bombing offered a
significant distraction of Luftwaffe assets and a substantial number of soldiers for flak
defense from the war with the Soviets, it did not serve as a major degradation of the
Reich’s industrial capacity (Magenheimer 1999, 234-235; Beaumont 1987, 6).

The bulk of the effort to defeat Hitler’s regime fell upon the hard-pressed Soviets.
And while Stalin had begun calling for a second front against German shortly after the
American entry into the war, these pleas went unheeded (Botj er 1996). Even with the
USSR in a precarious position throughout 1942, Roosevelt was too risk averse to send an
invasion force against Western Europe until he was nearly certain of success.

The executive appeared to believe that heavy casualties might undermine support
for war mobilization (Kimball 1996; Higgins 1970). A planned invasion of France in
1942 designated Operation Sledgehammer, was quickly dismissed as unrealistic, even in
1943 preparations for a continental invasion were not in place (Beaumont 1987, 5-7).
Instead the bombing of German controlled Europe constituted the second fiont for much
of the war.

Domestic political actors’ perceived utility of the issue-goods at stake in the war
was boosted by the realization of the scope of Nazi barbarism. In mid-1943 only half of
the respondents surveyed believed that the rumored death camps were true, a figure that
climbed to 77 percent at the end of 1944 (Mueller 1971, 375). Although the full scope of

the Holocaust was not revealed until at least 1945, it appears as if a growing sense of

 

17’ It was not until the Allies started targeting the oil production of the Reich, rather than the less focused
efforts at destroying urban-industrial areas, that a serious impact upon the Wehrmacht was realized
(Clodfelter 2006, 6).

179

moral outrage enhanced domestic political actors’ perceived benefits of destroying the

Third Reich.‘72

Figure 14: Lowess Graph of Anticipated War Duration and Win Ratio for
European and Pacific Theatre

 

 

30 H! Jan-44
‘ Aug-43
' Apr-13 Mar-44
2S _, Jan-42_ _.. g. Jun-43 Jun-43
Oct-43
Jun-42 Jan—43
20 Jun-42‘s. Jul-44
Anticipated War Apr _43 \Sep-44
Duration \ \ Oct-43
(Months) 15 \ Jan-44
Jan-43 A g_43\\Mar-44
i ‘\ Jul-44
10 ’1‘ Jan-4S
Jun-44 Aug-44
‘\ Nov-44
5 T Sep-44
er— mi- r . , .
0 '2 '4 -6 .8
Win Ratio
bandwidth = .8

When the long awaited D-Day invasion did come, it was well planned and
executed. Much of its success was owed to the Wehrmacht’s exhausting struggle with
the Red Army (Ambrose 1994). After the landings in June 1944 military operations
exploited German disadvantages in terms of equipment and numbers (Kennedy 1987,
353-357). Aside from the shocking Ardennes Offensive the Western Allies steadily

defeated their determined but vastly overmatched opponent.

 

172 A May 1944 survey indicated 81 to 10 percent of respondents favored unconditional surrender for
Germany (Public Opinion Polls 1944, 296).

180

While casualties were high in the European theatre from D-Day to VE-Day, these
losses did not provoke substantial amounts of political pressure. Even though Roosevelt
faced a more vigorous challenge than he had faced in any of his previous reelection bids,
his war policies did not substantially heighten his risk of deselection (Harnby 1992, 49).
The executive was comfortably reelected in November 1944.

Battlefield victories over both the German and Japanese armed forces led to
changes in the perceptions of the war’s utility. Despite enormous economic costs and
heavy loss of life the American public appeared to offset these costs by the benefits of
further fighting. Figure thirteen indicates total number of victories that the US had
divided by the cumulative number of battles.173 As the percentage of American wins
increased over its Axis opponents in both the European and Pacific theatres the lower the

perceived time until the war would be won.'74

The expected duration of the conflict
helped to improve the benefits of fighting by reducing the time discount until the
realization of benefits as well as reducing the expected costs of further fighting for the
American public.

Domestic political actors witnessed the realization of unconditional surrender of
Germany in May 1945. The greater difficulty would be the destruction of Imperial

J apan.175

The island hopping strategy had succeeded in grinding down Tokyo’s forces,
but the home islands remained unconquered. Despite increasing costs in terms of lives a

survey in February demonstrated that public support for Japan’s unconditional surrender

 

'73 See Chapter 3 for more details regarding coding rules of win ratio.

’7’ The perceived time takes the percentage of respondents beleving that the war will last a given amount of
time. The percentage of the population that holds this opinion is multiplied by the number of months and
these values are summed to indicate the perception of war duration for the aggregate population.

'75 The public fi'equently posited that the war in the Pacific would take much longer than had the war in
Europe, according to the average of anticipated war duration.

181

was 75 to 12 percent (Public Opinion Quarterly 1945, 95). Even with what seemed like
enthusiastic support for further mobilization, serious trepidation over the public’s

patience was aroused (Frank 1999, 343). '

Figure 15: World War II US Monthly Battle Deaths

 

 

20000 Dec-44
l J l-44 Apr-45
u
i Jan‘45Mar-45
l J .44 Nov-44
; un
15000 1 Sep-44 Feb-45
US i Aug .4 4 Oct-44
Monthly
Battle }
Deaths I
10000 a
May-45
May-44
e
Jan-44 Apr-44 Jun-45
5000 NOV-42 Jul-43 Nov-43Mar-44
T Jan-43 May-4% _4 Dec-43 Jul-45
May-42 Jun-42 Apr-43 "8 3Oct-43 Aug-45
l Dec-41 Jul-42 Feb-43 Sep-43
Apr-42 Aug-42 Dec-42Mar-43
i Feb-42 Oct-42 Jun-43
0 A Jan-42 Mar-42 Sep—42
T I T T I
0 10 20 30 40
War Duration (Months)

Despite having been steadily driven from its defensive frontier, Japan retained the
capability to extract costs in American lives (Slantchev 2003b). A fact that was vividly
demonstrated as the Allies neared the home islands. In the battles of Okinawa and Iwo
J irna a substantial but poorly supplied force of Japanese had to be dug out of their
fortifications at the expense of heavy US casualties. Starting in January 1945 kamikazes

were being employed to great effect against the US Navy.

182

In three months of fighting of the island of Okinawa the combination of kamikaze
attacks and fierce army resistance had cost 50,000 American casualties of which 12,000
were fatalities (Skates 1994, 76). These losses spawned apprehension as to the likely
number of casualties that it would take to secure unconditional surrender through an
invasion of Japan (Skates 1994, 74-82). Having recently replaced the deceased
Roosevelt, Truman now faced the task of continuing to garner the support of the
American public for firrther mobilization.

With the capture of the Mariana’s Islands, virtually all of metropolitan Japan was
within range of the B-29 high-altitude bomber. While the first bombing raid on Tokyo
from the Marianas took place in November 1944, it was not until January 1945 that the
efficacy of such attacks was perfected using incendiary munitions (Clodfelter 2006, 9-
10). Japan was also cut off from its sources of raw materials and other supplies by
intense submarine patrols along with the mining of the interior waterways of the home
islands (Skates 1994, 48-49).

Whereas these operations were comparatively low in costs, it was not evident that
these efforts at crippling the Japan’s economy alone would induce unconditional
surrender (Skates 1994, 50; Frank 1999, 35). As such the main thrust was still attacking
islands that could serve as logistical bases to support the scheduled invasion of Japan.
Projections as to the human costs of invading Japan ranged considerably but Operation
Olympic, which included only the invasion of the southern home island of Kyushu, had
estimated casualties that reached as high as 150,000 (Frank 1999, 138).

The mobilization of the entire Japanese population into soldiers with crude

weapons and the mass civilian suicides accompanying American conquests did little to

183

hearten the American public for an invasion (Frank 1999, 178-196). Because casualty
estimates were so high Chief of Staff Marshall and others military leaders actually
quashed the more disturbing casualty figures in order not to invite Truman’s suspected
casualty aversion from affecting his decision-making (Frank 1999, 133-148).

Before the most difficult decisions regarding an invasion had to be made the
Manhattan Project successfully tested its first fission weapon. Now an even more
devastating punishment strategy of atomic bombing could help to hasten the end of the
war, rather than the anticipated attack on Tokyo-Yokohama tentatively planned for July
1946 (Skates 1994, 215).

The justification for dropping the atomic bomb is a subject of intense debate. ’76
However it appears to be accepted that the heavy casualties anticipated in defeating the
Japanese significantly contributed to the decision.'" Employing nuclear weapons as part
of an effort to heighten the level of punishment does not correspond with political
pressure arising from the American electorate. However the modification of strategy is
undertaken out of anticipation for the possibility of deselection or reticence to further
mobilize in the future.178

Another effort to alter the anticipated costs of further fighting was the relaxation

of the demands for unconditional surrender. Some in the State Department had been

 

”6 Some scholars claim that it was almost entirely a function of intimidating the Soviets or considerations
of using the weapons that had been created at massive expense in order to determine their efficacy
(Alperovitz 1965). Scholars also point to wildly inflated casualty estimates that were used after the fact to
justify using the Atomic bomb. Other more even handed approaches stress the importance of compelling
surrender and Truman’s growing unease with Soviet behavior (Berstein 1975).

'77 For Operation Olympic, the invasion of the Japanese island of Kyushu an estimated 125,000 casualties
were anticipated with roughly 25% of these as deaths (Skates 1994, 79). However, relatively favorable
battle conditions might be greater when launching a follow up attack upon the Tokyo-Yokohama area.
Perhaps another 13,000 fatalities from kamikaze attacks as the American fleet approached the home islands
(Frank 1999, 187).

m In fact opinion polls conducted immediately following Japan’s surrender indicated that 85 percent of
respondents supported using the weapons, vastly higher than the number of dissenters (Frank 1999, 331).

184

given the impression through MAGIC (the Allied system of reading Japanese diplomatic
transmissions) that there was a substantial peace movement amongst domestic political
actors in Tokyo (Toland 1970, 761). After the destruction wrought by the nuclear bombs
and the Soviet invasion of Manchuria, the reduction of issue-goods negotiation position
to accommodate a role for the emperor by Truman indicated an effort to stave off further
costs.

World War II represents a case in which the public did not agitate for changes in
war policies. The major fear that the public might fail to mobilize came from the
anticipated duration and human costs associated with toppling Imperial Japan, especially
as their armed forces adopted fanatical strategies designed to inflict maximum human
losses. Anticipating the prospect of political pressure rather than political pressure
helped to inspire a change in military strategy and modest changes in issue-goods
position.

World War H constitutes a conflict in which the public ascribed much greater
benefits to successfully winning the war than they did in Vietnam and Korea. Yet, it also
appears that war policies were undertaken with the intention of maintain public support
for mobilization. Further, credible evidence as to the likely outcome of the war was
achieved through the successful defeat of the Axis in battle. The probable defeat of
Germany and Japan was interacted with the stated objective of unconditional surrender

increased the appraised benefits of the war.

185

Conclusion

The principal of convergence employs casualties, military reserves, and battles
serve as objective indicators of the likelihood of prevailing in a war or at least that the
executive aggregates the preferences of the domestic political actors and on that basis
forms war policy. Paralleling this assumption is the finding that casualties are a major
source of dissatisfaction with war policies which implies an infra-war assessment by the
democratic public. Assuming that a unified interpretation of battlefield events is made by
both executive and domestic political actors, than the executive would withdraw from the
conflict in concert with public opinion as costs outstripped benefits.

Instead the differentiated opinion of a war’s utility between domestic political
actors and executive is based upon an intra-war assessment. For the public updates in the
perceptions of costs and benefits interact with characteristics such as partisanship
eventually take the place of initial rally effects in fashioning public perceptions of a war
(Feavar and Gelpi 2004, 149-164).

An executive who extols the public to support further mobilization still needs
credible evidence of success in order to secure public support. Support for a war can be
bolstered by changes in war policy, such as articulating demands for unconditional
surrender during World War II or the Linebacker bombing campaign during the Easter
Offensive. As with the autocratic cases the public appears to move toward a less
enthusiastic endorsement of the war over time. Although casualties and war duration
both appear to lower the relative utility of a war, this is substantially offset by the benefits

of fighting.

186

Reiter and Stam find that the longer a war goes on, the public refuses to mobilize
and the less likely a democracy is to win (2002).179 However an executive might be able
to anticipate the dwindling willingness of the public to mobilize and adjust war policies
accordingly.

Explored in greater depth in the following chapters, it appears that both issue-
goods benefits and military strategy are both changed in an effort continue to garner
support for mobilization. Doubtless bureaucratic politics pervades in the executive and
tactics are changed in accordance to key variables, at a more macro level of analysis
negotiation position and military strategy are fiequently driven by political pressure.
What emerges is a new finding that instead of the behavior of states being conditioned by

regime type, it is what is assessed as costs and benefits that differs.

 

”9 I find that there is not a strong relationship between a democracy’s percentage of wins and the duration
of a war (correlation 0.0789). Moreover, OLS models appraising this relationship were likewise
insignificant.

187

Chapter Four: Change in Issue-Goods Position

As wars progress it is found that domestic political actors examine the costs and
benefits of an executive’s war policies. If domestic political actors come to view war
policies as significantly inferior to plausible alternatives they will apply political pressure
to instigate a change in how the executive is conducting the war.

While the case studies appear to conform to the pattern of political pressure
motivating changes in war policy, statistical tests will be employed in order to determine
if this correlation holds when alternative hypotheses are controlled for. Also, the relative
importance of different variables with respect to the likelihood that strategy and issue-
goods position will change is also assessed. If the weight of domestic political pressure
when compared with alternative sources of changes in negotiation position is
comparatively trivial, then the more parsimonious assumption of a unitary assessments
would likely be a superior theory.

Assuming that states acquiesce to fighting in accordance with a comparatively
favorable cost-benefit assessment, this understanding as to what a state can obtain is
thought to change as the a war progresses. The ongoing bargaining model of war
contends that states will credibly reveal their relative balance in terms of capabilities and
resolve and that a negotiated settlement will be made that reflects this distribution of
martial ability. If the principal of convergence is strictly correct than battles, casualties,
and other evidence of the relative resolve and capabilities of the combatants should
dictate changes in a state’s negotiation position eventually culminating in a termination of

hostilities.

188

The principal of convergence literature is largely reliant upon casualties data as
the impetus behind why states eventually achieve a negotiated settlement. Specifically as
a state’s soldiers are killed in greater proportion to its population than does its opponent
this will eventually compel a state to cede for peace under less-favorable circumstances
(Slantchev 2003a). However this data is of dubious validity in its application to the
question of the principal of convergence.

Specifically the importance of casualties as a percentage of the population is only
relevant in that most wars do not continue to the point of draining a state’s manpower to
the point of exhaustion. Further, the relative importance of casualties is markedly
different on the basis of regime type, with democracies displaying a much greater
sensitivity to human losses. I propose to use metrics that should offer greater
measurement validity in evaluating the principal of convergence.

A measure of success and failure in war that is endogenous to the issue-goods at
stake in the conflict is employed. By using battles instead of casualties and population,
an appreciation for the relative success of the combatants can be made. The combatants’
relative level of mobilized resources and the objectives being sought after are
contextualized by the outcome of battles. In the Vietnam War American manpower
reserves where never substantially diminished by battlefield casualties. Similarly despite
heavy losses, the Chinese did not ever have too few men to potentially field in their fight
for Korea.

By contrast the issue-goods being contested are more adequately encapsulated by
the metric of battles. German military victories secured territorial acquisition and the

political collapse of their opponents. The American engagements in Vietnam were

189

designed to destroy their opponents and reduce the insurgency’s potency. The territory
acquired in the Iran-Iraq War indicated the comparative balance of military capabilities
and the resolve of the opponents to continue to mobilize, exemplified most clearly by
Iran’s htunan wave tactics. Contextualizing the political dispute at the center of a war by
examining battlefield victory and defeat, an indication of how likely the antagonists are to
achieve these aims is established.

Employing a new measure of battlefield results hypotheses can be evaluated as to
what inspires a state to alter its negotiation position and eventually to terminate a conflict.
Amongst these hypotheses the validity of political pressure can also be tested controlling
for the objective variables of military defeats, casualties, war duration, and other factors.

The following section details the concepts employed and the creation of variables.

Issue-Goods Negotiation Position
A state’s preferences in terms of what settlement could be reached to end
hostilities. The major problem with deputizing negotiations as the dependent variable is
that these offers might be considered cheap talk without consequence. If negotiations are
only so much posturing, then analyzing the correlates of bargaining offer is unprorrrising
means of testing the principal of convergence. ’80
A state can misrepresent itself in terms of its ability to garner more in the

distribution of the issue-goods at stake in a war (Wagner 2000). Given that war is a

negative sum game, insincere negotiations would have costs associated with a diminished

 

"’0 Some of the principal of convergence research has latched onto negotiations as informative events. The
screening hypothesis of Powell examines negotiation offers as an important feature in understanding the
relative strength of an opposing state (2004). Slantchev also identifies negotiations as an informative
means by which to identify the comparative resolve and capabilities of an opponent, but negotiation offers
were also found to be less credible than the information provided through fighting (2003a).

190

utility endogenous to further fighting. More importantly, the possibility of jeopardizing
future negotiations with an enemy state through disingenuous negotiations, might serve
as a cost that bolsters the validity of negotiation offers (Fearon 1998). Additionally the
executive of a state makes negotiation offers observed by domestic political actors who
anticipate a settlement, non-credible offers might disappoint these expectations
engendering audience costs (Schultz 1999).

Even with these limitations an analysis of negotiations has validity for the purpose
of studying what compels states to alter bargaining position. Negotiations offer one of
the few observable implications as to what a state believes it can garner in issue-goods
benefits previous to the conclusion of a war. Bargaining offers made will frequently be
more optimistic than what would reflect a state’s true appraisal as to the issue-goods
benefits it is capable of garnering. In the course of a war a state might posture itself more
capable than it actually by demanding greater concessions in a settlement. While
negotiation offers might not reflect a state’s true beliefs as to its relative resolve and
capabilities, gradual changes for concessions or demands made in a series of bargaining
offers can reflect changed expectations.

My dissertation contends that domestic political actors are capable of intra-war
updates of the costs and benefits of a war. If a collection of domestic political actors
assess the utility of current war policies as being considerably less beneficial than some
alternatives, this should result in the application of political pressure. Assuming that '
domestic political actors ascribe less utility to current war policies than does the
executive, this might compel changes in military strategy, reduced mobilized resources,

and/or alteration of issue-goods negotiation position.

191

benefits (probability of success X utility of security goods) / time discount > costs (casualties + other costs)

A reduction in the issue-goods demanded from the opposing state could enhance
the likelihood of achieving the more limited issue-goods benefits. An example would be
the removal of Beijing’s demands for the Chinese UN Security Council seat in return for
an armistice in Korea. The lowered utility of the issue-goods position is offset by an
increased probability of an opponent accepting the settlement.

A reduction in the issue goods position of a state might also have the effect of
reducing the perceived amount of time that is likely to transpire before a state will realize
the benefits of a war.’81 Nixon’s decision to drop demands for the withdraw of NVA
troops from South Vietnam reduced the anticipated time until an agreement was reached
with Hanoi. As the issue-goods position of a state is reduced this generally reduces the
level of resources that are mobilized to fight the conflict, even short of a termination of
hostilities.

A similar concept to the time discount of benefits is the costs incurred in pursuit
of issue-goods benefits.182 Domestic political actors’ assessment of costs relative to
benefits informs as to the utility of current war policies. The economic opportunity costs
of China’s intervention in Korea were more dear to some domestic political actors than to
Chairman Mao. What constitutes costs and the relative weight that one assigns to costs

can be substantially different between the executive and domestic political actors.

 

"" Pillar illustrates that denial of the objectives of a war are a cost that state’s appreciate when deciding to
intiate negotiations and conclude a conflict (1983).

"2A problem with any cost-benefit analysis conducted in an ongoing war is that costs are accrued relatively
quickly. Lives lost, money spent, opportunities forgone, are realized fairly quickly. Commensurate
benefits of security, deterrence, credibility, prestige might not be realized at the same rate.

192

Negotiations might offer a way for the executive to reduce anticipated costs
through the pursuit of a more manageable objective. As evidenced by the United States’
modification of its demands for unconditional surrender helped to mitigate the need for
an invasion of Japan. I theorize that the greater the amount of domestic political pressure
the greater the probability that an executive will make more conciliatory negotiation
offers in terms of the issue-goods at stake in the war.183

A negotiation offer is defined as an overture made at the behest of the state’s
executive to its opponent stating acceptable conditions for war termination. An
illustration of an un-coded negotiation offer would be the Nazi officials’ late war
attempts at bargaining that were not authorized to by Hitler himself. Whereas a peace
offer to Moscow in return for an independent Ukraine and material resources to continue
the war in the west, is coded in that they reflect the wishes of the state’s executive
(Erickson 1975b, 162).184

Three categories of offers are designated. The first of these category, labeled -1,
are settlement offers that would give the bargaining state less of the issue-good at stake in
the war than they possessed at a conflict’s outset. The second category 0 indicates that

within that month a state did not make any offer to their opponent. Category 1 indicates

that an offer made by a state approximates the distribution of issue-goods at the onset of a

 

"’3 Another possibility is that issue-goods position is increased in order to mitigate anticipated domestic
political pressure. Goemans notes that an executive fearing future sanction might try to buy off domestic
critics through side payments obtained as spoils of war (2000). The possibility of a state increasing
demands of the enemy, taking a risk acceptant stance, in order to secure greater issue-goods benefits is a
theoretically plausible means by which negotiation position might be used to address political pressure.
This again presumes that diversionary aims or poor military execution might result in post-hoe deselection.
By contrast 1 do not find evidence that heightened domestic political pressure brings about negotiation
offers with heightened demands for issue-goods.

134 The reasoning behind this coding system is that the state’s leadership is able to make determinations of
battlefield information and then respond to them. Whereas unsanctioned efforts to mediate a conflict
represent discontent with war policies, which is a phenomenon specified as an independent variable.

193

war. Finally, category 2 indicates that a state is making an offer that grants them more of

the issue-good in question than they possessed at the onset of the conflict.

Table 2: Frequency of Negotiation Offers

category frequency Percentage

 

 

0 616 81.7%

1 113 15.0%

-1 9 1.2%

2 16 2.1%

Total 754 100%

As illustrated by American policymakers who sought a viable anti-communist ally
in South Vietnam, whereas the North Vietnamese sought reunification. The negotiations
between both states throughout the conflict are coded as category 1 events, in that the
negotiating positions of the states would have left South Vietnam in roughly the same
position as it was at the beginning of direct American involvement in the war. By
contrast, the Allied proclamations of unconditional surrender made in World War II are
coded as category 2 bargains and conversely Japanese counter-offers to secure the role of

the emperor are coded as -1 category offers.

Political Pressure

In order to measure political pressure two different sets of variable are used
contingent upon the regime type. The key casual link pursued here is that negative
appraisals of war policies made by domestic political actors will increase political
pressure upon the executive for war policy changes or enhance the likelihood of

deselection.

194

For the democratic cases the level of political pressure is defined by the job
approval percentage of the executive. Presidential or prime minister job approval offers
some advantages over more narrowly cast questions such as support for a war, but also
have limitations. Policies and events unrelated to war can negatively and positively
affect public approval, most notably perceptions of the economy (Kramer 1971).
Discontent associated with scandal or with the general erosion of popularity throughout
the term in office can also influence public approval.

Presidential approval elicits responses that directly address the president’s
standing. The institutional literature stresses the ”importance of the fear of losing office as
the principal driver in many executive decisions (Bueno dc Mequita et al. 2003). If this is
the case, than perhaps no periodic intra-election metric is more accurate in determining
the likelihood of sanction than presidential approval.”5 If de-selection and/or the loss of
political authority are discernible to politicians, I assume that presidential approval is a
highly attenuated measure for the likelihood of being deselected from power or losing
political clout.“’6

Public opinion data reflecting the job approval rating for a state’s executive is the

base measure. The level of public support for the executive is discounted by the amount

of time until the next scheduled election.”7 To illustrate a September 1965 Gallup

 

"’5 The United States Congress is found to be susceptible to electoral sanction as casualties in their
constituency increase (Cotton 1986; Gartner, Segura and Barratt 2004). Such that even those not directly
linked to foreign policy are subject to punishment helps illustrate the connection between approval for a
conflict and presidential sanction.

’“Canes-Wrone and de Marchi (2002) find that levels of presidential approval are at least partially salient
in explaining the likelihood of legislative success in Congress.

"’7 The public opinion surveys of the United Kingdom are likewise discounted fi'om the time to the next
election. While obviously Britain’s parliamentary system does not have fixed election dates, the incumbent
party being able to establish the time of elections can reasonably be assumed to understand the relative
irnminence of the upcoming election. Consequently the governing party will presumably have an

195

survey found 53 percent of respondents approved of the job President Johnson was doing
in office. The figure of 53 is divided by 39, the number of months until the next
elections. The result is then logged (natural log) in order to mute the effects of extreme
values.

The rationale for the stress on time until election is that in a democracy elections

are the principal arbiter of deselection.188

An understanding of current job approval
ratings informs as to the likelihood of maintaining office, but the security or insecurity
implied in the level of popular support is most relevant at the time of the election. The
expectation is that an executive with high approval ratings nearing election is
comparatively secure, whereas an unpopular leader or an executive with strong support
but with a greater amount of time until the election is comparatively less secure.

The measure of political pressure in the autocratic cases makes reference to the
effectively enfranchised portion of the population. These actors are typically individuals
with positions of influence in the bureaucracy, dominant political party, military,
terroristic police/repressive apparatus, or are members of a significant political institution
such as the Politburo. Despite state domination of the media these individuals are able to

make calculations with respect to the costs and benefits of an ongoing war and are

knowledgeable of alternative war policies they would prefer.

 

understanding of the relationship between this date and the current trajectory of public opinion with regards
to the likelihood of losing office.

"’8 By contrast in an autocratic state it is not especially clear when discontent with executive policies will
result in deselection or punishment. Accordingly, the importance of changes in an executive’s public
approval rating in a democracy are discounted by the amount of time until the election is at hand thereby
stressing the comparative imminence of being deselected form power. All surveys come fi'om Gallup
Polling data from all three wars. In the rare case in which a survey is not taken in the intervening month
the change in presidential approval is averaged for both the current month and the month prior in which no
survey was taken.

196

Domestic political actors can potentially influence war policy because they have
the ability to deselect the executive or reduce his political authority. Whereas in a
democratic context the enfranchised citizen has a single vote with the majority or
plurality deciding the election, the influence of domestic political actors is not as
homogenous. Leaders of the military and repressive apparatus likely have a greater
influence in deselecting the executive than an economic minister.

Autocracies do not have clear decision rules determining the selection of the
executive. The sufficient conditions for deselection of the executive is only determined
after it has been accomplished, the influence of a single individual’s application of
political pressure is therefore unknown. A further difficulty in specifying political
pressure in an autocracy is the potentially high costs of expressing dissatisfaction with the
executive. Undertaking measures to deselect the executive or even advocating policy
alternatives can be very hazardous.

The measure of domestic political pressure must account for the plausibility of
deselection as well as capturing the latent opinions of domestic political actors. Such that
observable events that reflect high levels of dissatisfaction with executive policies are
referenced. Even those not party to political pressure events, might share the
comparatively unfavorable opinion of current war policy, but are not inclined to
undertake the costs themselves.

Political pressure events are coded as a dichotomous variable for each calendar

month of a war.189 Four different events were specified, purges, coups, assassinations

 

"’9An attempt was made to create an index that included different values for the different political pressure
variables, however the relative rarity of certain incidences made this problematic. Also a measure of
autocratic political pressure used a time discounted formula in which older pressure events were weighted

197

and executive removal. If any of the events occurred in a given month then a l is
recorded otherwise 0. A cumulative tally of these events is then created with the number
of political pressure event-months added to one and then logged. Once the specter of
deselection becomes apparent more pressure event-months should have a diminishing
impact in informing the executive as to threats to his/her rule. ’90

A difficulty with utilizing any of these four variables is that within autocratic
regime’s these events occur outside of the context of a war. An examination as to the
motivations of these political pressure events utilized in my analysis finds that war
policies are very fiequently the causal impetus behind the recorded manifestations of
political pressure or at the very least war policies exacerbate existing tensions.

The purge variable indicates that political elites have been eliminated from their
positions of power. A sliding scale of events is used to identify purges, a small number
of high ranking officials or a larger number of progressively less influential actors.
China’s party-wide purges, Hanoi’s dismissal of a member of the Politburo as well as the
incarceration and execution of high-ranking Soviet Army officers are coded identically as
purges.

Purges, obviously, are repressive events and indicate that the executive is still
well within their capacity of maintaining power in spite of political pressure.
Acknowledging the distinction between mass repression and repression of elites, a purge
implies a challenge directed at the leader rather than the regime itself. Purges might then

serve as a deterrent designed to curtail latent discontent from becoming manifest in an

 

less than more recent events. However, this explanatory variable performed poorly and did not appear to
reflect the executive’s lingering apprehension once some political pressure became evident.

mln order to control for the level of repression that a state undergoes the autocracy variable from J aggers
and Gurr was utilized (1992). It did not appear to have an appreciable effect upon the political pressure
variable.

198

attempt to supplant the executive. A purge can be costly to a state, expunging military
and/or bureaucratic expertise, lowering morale amongst elites, reducing policy innovation
and autonomy. 19’ Moreover, correctly identifying and repressing actual dissenters
amongst domestic political actors is difficult and increase these inefficiencies.

A coup is coded when an actualized attempt to seize power away from the
executive is made by a group of domestic political actors with a credible chance of
supplanting the leader. A prominent example would be the assassination-coup attempt
made against Hitler in July of 1944. The group that undertook this attack not only
planned to take the Fuhrer’s life, but also initiated a rudimentary plan to form a new
government to be installed once rival claimants were eliminated.

Likewise, assassination events are coded with a similar reference to plausibility.
An assassinations attempt is only recorded if the target was the executive himself and if
tangible steps had been taken to conduct the operation, rather than just referencing a
conspiracy. This is designed to filter out the actions of lone individuals or attempts of
dubious plausibility, along with having an observable reference point as opposed to
relying upon latent opinions.

Accordingly, the two conspiracies hatched to eliminate Hitler in Paris were not
coded in that he never visited on the occasion that he was to be murdered and no weapons
were procured or tangible efforts undertaken to execute the plan (Duffy and Ricci 1992,
18-19). Whereas attempts that included a bomb placed on the Fuhrer’s airplane and a
detonation inside of a conference room in which Hitler was present were both coded,

even though unsuccessful.

 

'9' Ambrose (1994) contends that the comparatively rigid autocratic structure of the German armed forces
in World War [I contributed to Allied success during the Normandy landings.

199

While both assassinations and coups are expressions of discontent with the
executive within the limited array of tools available to actors within an autocracy, the
removal of the executive is the successful expression of this discontent. Removal is
coded if the leader of a state is replaced during the war with a different leader, with or
without the acquiescence of the chief executive in question. While Tojo complied with

an effort to remove him as executive, this is still recorded as a change in leadership.

Battles

It is theorized that as a war is fought the relative strength and resolve of
opponents can be revealed through the costly interactions of battles (Wagner 2000;
Slantchev 2003). Battles are a quantifiable metric that transcend particular cases,
informing as to the likely duration and the probability of achieving the sought after issue-
goods as states demonstrate their relative military abilities.192

Unlike the principal of convergence literature, I would suggest that the executive
is not the only person privy to the information generated through fighting. Likely
autocratic elites have the expertise in military affairs or have access to such an
analysis.193 As these domestic political actors observe the outcome of fighting they can
assess the comparative utility of the executive’s war policies. Many of Japan’s civilian
and military leaders began to pressure the executive for a change in policy after the
Imperial Navy’s defeat at Midway (Butow 1954, 14). Or after the disaster at Stalingrad,

many German officers believed Hitler’s military strategy and micromanagement were

detrimental to the war effort (Hoffman 1988, 109).

 

’92 Slantchev (2004) uses casualties relative to reserves as the principal metric for demonstrating the
principal of convergence empirically.
93 This is borne out in many of the case studies in chapter two.

200

In the democratic cases it is difficult to imagine that the public has access to and
can digest the substantial significance of every engagement. However, significant battles
and military campaign have had an appreciable effect upon assessments of an ongoing
war. The capture of Atlanta prior to the Union’s presidential election helped bolster
public perception of the prospects for winning the Civil War (Dudley 1932). The
shocking battles of the Tet Offensive provide another ready example of the significance
that these engagements can have in fashioning public opinion (Blood 2005). Both high-
profile battles and a number of less visible battles in the aggregate are potential sources
of information from which the public can appreciate the utility of a conflict.194

A battle is coded if an attacking force, composed of at least one-percent of their
total combat forces deployed in that theatre, is engaged in deliberate combat operations
against an opponent. '95 Battles are distinguished fi'om each other in that they must be
either chronologically or geographically distinct. In the case of a major military
campaign individual battles are separated out of the larger operation.196

To further illustrate this coding system the following example is utilized from the

multi-month Stalingrad campaign of World War II. An initial German attack drove into

 

'94 A finding conditionally supported by Boettcher and Cobb in experimental data (2006).

’9’ The term theatre is meant to differentiate fi'om a nation’s total combat forces if they are engaged in a
war against two opponents, for example Nazi Germany against the USSR and Anglo-American forces or
the US fighting both the Germany and Japan. The 1% figure is made in reference to numbers of combat
soldiers involved in a battle or to the size of the formations involved, such as “division” “battalion”. When
evaluating the 1% criterion this number references all combat soldiers from different nationalities serving
under a unified command. For example an assault by US Marines undertaken in Vietnam during 1965 was
not coded as a battle, as this battalion sized operation did not constitute 1% of the “fine world” combat
forces under American command. As such every incident in which an attack was made by a group
consisting of at least 1% of their combat forces made at an independent military target counts as a single
battle.

"6 The rationale for utilizing a percentage of a military force as opposed to a criterion of a fixed number
troops, such as a division, is that some wars involve considerably smaller forces than other conflicts. As
such the threshold of what is a significantly informative engagement to the states involved may be much
lower than is the case in major wars. If a battle takes place continues across calendar months then it is the
month in which the battle is concluded that it is coded.

201

Stalingrad and captured portions of the city. A second German thrust aimed at capturing
the eastern industrial portion of the metropolis stalled short of its objectives. Another
German attack on this same area was later successfirl. This was followed by a major
Soviet encirclement operation undertaken in the area surrounding Stalingrad turning both
of the German flanks. The Wehrmacht’s attempt to rescue their encircled comrades also
failed. An initial Soviet attack upon the trapped German forces was beaten back but a
final assault broke German resistance within Stalingrad, concluding the campaign with

the surrender of the trapped 6th Army.

The criterion of separating out different battles from one another is also made
with respect to time. During the Vietnam War the fighting for the city of Hue took place
over two months, the NVA took the city in a successful attack, and was later ejected by a
counterattack of combined US-ARVN forces. These battles are coded as separate events
in that a chronological break in the fighting occurred even though fighting took place at
the same location.

Battles are also assigned a value of a win, lose or draw. This distinction is based
upon if the tenitorial objective under attack exchanged hands at the conclusion of the
battle. If the area attacked by one antagonist was captured, this results in a win for that
side and a loss being recorded for the opponent. If a battle is undertaken and no
substantial gains were made with respect to the territorial objective then this results in a
draw being recorded. This metric identifies not only success at acquiring territory at the
expense of an opponent, but also that the referenced state enjoys the military initiative.

For example the objective of the Allied landing at Normandy was to secure the

beachhead and the adjacent inland territory, this objective was achieved by 1% their

202

combat forces resulting in a win being recorded for the Allies. A contrasting example
would be the German relief effort to rescue their comrades in Stalingrad. Although the
6th Panzer was able to gain territory at the expense of the Red Army, the Germans still
fell well short of their operational goal and as such the battle is coded as a draw.

A variable for surprising victories and defeats was also created. Battles such as
Pearl Harbor, portions of the Tet Offensive as well as initial phase of the German
Ardennes Offensive are all indicative of engagements in which one of the opponents
performed more adequately than expected. The unexpected victories and defeats
variables are specified as the number of shocking losses and wins occurring in the
previous month and the cumulative number of unexpected victories and defeats from the

beginning of the war until the previous month.197

Hypotheses Tests

The principal of convergence literature contends that as a war progresses the
combatant states better understand the relative distribution of capabilities and resolve. A
war ends with a mutual agreement as to the ability of the states to achieve the issue-goods
in question. The expectation is that as the net utility of continued fighting decreases a

state should make less favorable bargaining offers.’98 If political pressure has no effect

 

197 In reference to the Pacific Theatre of World War II, naval battles are also recorded, always as draws, if
1% of the combat ship tonnage of the initiating force was present. However unexpected naval victories for
the antagonists are still recorded.

mA number of hypotheses tests were conducted based upon a survey of the literature. For a detailed
treatment see Bennett (2003). The screening hypothesis which holds that as a state fights battles create
credible information may reveal its true nature and inform the opposing state as to what kind of enemy they
face (Smith and Stain 2003; Filson and Werner 2002). This was evaluated taking the log and square of
both the cumulative number of battles and the cumulative number of defeats. The number of casualties
that a state absorbs as a percentage of its population and the duration of a war were also evaluated
(Slantchev 2003a). One hypothesis holds that states which have executed a number of strategies
throughout the course of a war is less likely to make bargaining offers (Bennett 2003). In that these

203

then the unitary actor assumption is essentially validated and alternative variables should
have the largest impact on the likelihood of negotiation offers.

The tradeoff implied is that the intra-war assessment of a conflict’s utility is less
parsimonious then the executive assessing costs and benefits in accordance with a states
preferences. However if a substantial portion of the variance in explaining changes in
bargaining position are attributed to domestic political pressure then this would be worth
the loss in a theory’s simplicity.

The first battlefield hypothesis is based on Smith and Starn’s model that stresses
the importance of the ultimate defeat of one of the antagonists (2004). The expectation is
that as a state’s ability to resist complete defeat is degraded they will make progressively
less advantageous negotiation offers. The total number of defeats that a state suffers up
to the previous month is indicative of the likelihood of a state’s eventual defeat.

H1: Cumulative Defeats should result in a state being more likely to make status
quo or sub-status quo oflers.

Wittman contends that while battles may inform they nright also simply compel
the two antagonists to reevaluate their issue-goods positions (1979). A sequence of
defeats incrementally realigns the Victor’s expectations upward and their opponent’s
downward. These progressive revisions in expectations of a war’s utility might not be
sufficient to inspire negotiation offers that are likely to be accepted by their opponent.

Instead military defeats that radically alter previous expectations as to the relative

ability of a state to garner issue-goods benefits should be much more influential than

 

relatively dexterous states consider themselves to be better able to handle adverse military setbacks. the
number of strategy changes and negotiation offers trend upward together as an executive attempts to
mitigate domestic political pressure through both means. Alternatively, it is possible that previous strategy
changes would improve the cost-benefit calculations of domestic political actors, reducing the imperative
of latter negotiation offers. In any case strategy changes were insignificant (strategy count and log of
strategy count), especially in the Heckrnan probit selection models.

204

anticipated battlefield outcomes. A shock to previous assumptions regarding the balance
of capabilities and resolve influences a state to accept a more reduced issue-goods
negotiation position. Two variables are specified one for unexpected losses in the last
month and the total number of unexpected losses.

H2: Shocking defeats should result in the defeated state being more likely to make
status quo or sub-status quo ofi‘ers.

A state can likely assess its ability to obtain a given set of issue-goods on the basis
of its record of victories and defeats (Wagner 2000; Slantchev 2003a). The comparative
efficiency that a state enjoys in terms of battlefield outcomes is another key metric by
which the likely outcome of a war becomes clearer. Enjoying the military initiative and
succeeding in attaining the territorial goals are both indicative of a superior military
capacity.

The variable win ratio is the total number of battles won divided by the total
number of battles fought. A state’s change in win ratio was also specified using the
previous month’s winning percentage subtracted by the winning percentage from two
months ago. A recent decline in a state’s relative military success could help to catalyze
a lowered issue-goods negotiation offer.

H3: States with a lower winning percentage of battles fought should be more
likely to make status quo or sub-status quo bargaining ofi’ers.

Often industrial strength is important factor in determining military success (Stam
1996). The relative balance of capabilities also helps to establish previous expectations
as to a war’s outcome. ’99 The COW’s capabilities index is utilized, with the reference

state’s material capabilities divided by its opponent’s (Small and Singer 1982). Another

 

l9”Biddle (2004) illustrates the mixed effect that capabilities have on the outcome of battles and wars
generally. They suggest that overall level of capabilities is relatively unimportant in determining the
outcome between states.

205

variable is also specified which shows the change in the balance in capabilities over the

course of the war.200

The difference in win ratio variable takes the initial capabilities

ratio and divides this by the cm'rent capabilities ratio. The growth in the discrepancy of
material capabilities between combatants should result in an increased likelihood of less
advantageous bargaining offers (Slantchev 2003b).201

H4: A weaker or weakening state is more likely to make status quo or sub-status
quo bargaining oflers than a strong state.

Battlefield fatalities have been used in studies evaluating the principal of
convergence as a metric indicating the likely outcome of a war (Slantchev 2003b;
Ramsay 2006). A state that is taking more casualties and has a smaller population is more
likely to be ultimately defeated. The greater these losses the more likely that a state will
make conciliatory negotiation offers. The cumulative number of monthly casualties a
state has suffered is included as a metric of costs. The ratio of monthly casualties
between states was also specified, taking the reference state’s losses divided by the
opponent’s battle deaths (Boettcher and Cobb 2006).

H5: A state that has experienced greater human costs throughout a war may be
more likely to make status quo or sub-status qua oflers.

As an executive feels a dissipation of political authority or an enhanced hazard of

deselection, this should increase the likelihood of an alteration in war policies to improve

 

2°° The capability ratio between states is subject to change from year to year, but also as minor allies enter
or exit the leading state’s coalition. These same coalition of states were also referenced for the casualty
ratio variable. For example the defection of Italy and Finland, are calculated as reducing Germany’s level
of capabilities in the months following their departure.

Minor states included are Italy, Hungary, Romania and Finland for Germany in WWII. Australia and New
Zealand for the US in WWII on the Pacific Front. The UK for the US in the European theatre of WWII.
The South Koreans and South Vietnamese for the US in those respective conflicts. The UK for France in
WWII.

20’ For the purposes of numbers of battles and beginning capabilities ratio only if one of the states is
completely defeated, France in WWII and North Korea in the Korean War, do these variables reset against
their new opponent. Otherwise an ongoing count of battles and the initial balance of capabilities are
referenced.

206

the utility of his/her domestic political actors. The executive might change a state’s
issue-goods negotiation position in order to increase the probability of a lesser issue-good
benefit, reducing the time discount that domestic political actors might place on future
benefits, and/or decreasing the anticipation of future costs.

H6: An executive facing increased evidence of political pressure is more likely to

make status quo or sub-status quo oflers.
Analysis

A number of methodological options were employed to evaluate the above
hypotheses. The first method applied was a multinomial logit analysis. While a ranked
interpretation of the negotiation offers can be made, ordered and stereotype logit models
both had consistent violations of their regression assumptions. Theoretically it is also
unclear if no negotiation offers are superior or inferior to making a status quo offer to an
opponent. For example an executive’s utility may be enhanced by a diversionary war for
some period of the conflict, but then later wants to settle (Ostrom and Job 1986; Levy
1989).202

In order to better understand which negotiation offers are associated with
favorable events and which are associated with negative events, multinonrial logit models

were specified.203 All observations are referenced against no offer being made

 

202Acknowledging this possibility the analysis makes no universal assumptions regarding the relative
superiority of l as opposed to 0 bargaining offers, although broadly speaking adverse battlefield
information and political discontent promote bargaining offers.

203 Pooling all observations together might also be inappropriate. States initiating conflicts are
hypothesized to have made a positive cost-benefit calculation prior to beginning a war (Bueno de Mesquita
1980). A greater amount of disconcerting information in terms of costs and benefits might be needed in
order for the attacking state to seek an intermediary settlement. If both initiating and defending states are
tested together this might negate the importance of some explanatory variables. A rare events logit test of
just the initiating states found that the cumulative number of defeats and the change in capabilities ratio are
the only significant explanatory variables. The initiating states are Germany, Japan, Iraq, North Korea-
China, North Vietnam.

207

consequently neither the democratic or autocratic political pressure variables were
specified. These models also included the logged number of months since the last
negotiation offer made by the state and the logged number of negotiation offers made up
until that point?“

Another pair of controls were also specified, first war duration is controlled for by
logging the number of months over the course of a conflict. It is reasonable to assume
that the greater a war’s duration the greater the likelihood of negotiations in that wars
eventually end, however this variable was routinely insignificant. A second variable was
employed to control for reciprocation of another state’s negotiation offers (Axelrod 1981;
Pillar 1983). Opponent negotiations are operationalized by taking the most recent month
containing an opponent’s offer and dividing one by the number of months that have
passed since the last offer was made (l/t).205

Very few of the independent variables testing hypotheses one through five were
significant in multinomial logit models. Many of the models specified had violations of
the independence of irrelevant alternatives assumption when multiple explanatory
variables were evaluated simultaneously. Another problem for the multinomial logit
models is the comparative rarity of both category -1 and 2 offers.

A potential difficulty is that by pooling all observations together both the

initiating and defending states are assessed simultaneously. Given the expectation that

states select war they think they can win (Bueno de Mesquita 1980), the impact of

 

20" Squaring time and an unmodified negotiation count variable were also specified, however these did not
perform as well across specifications. Additionally logging these variables appears to be more in line with
expectations of the effect of time and previous negotiation offers, some cases saw a large number of
negotiation offers. These logged variables are scaled by adding 1 to the initial value and then logging the
resulting integer.

20’ The military expenditures of a state were also specified in order to control forthe costs of a war, but
this variable was not statistically significant at conventional levels.

208

battlefield events is plausibly greater for the initiating states as they alter their previous
beliefs more than do their targets. When both types of states are tested together the
significance of explanatory variables might be negated if previous expectations of the
utility of fighting have previously optimistic initiators revising expectations downward.
Simultaneously the less optimistic defending states enhance their costs and benefit
assessments as they exceed previous expectations.206

Because sub-status quo offers being made by the initiating state should be the
most dramatic reevaluation of a war’s likely outcome a binary analysis of categories -1
against no negotiation offers is made. A complication associated with using a
dichotomous hypotheses test is that selection bias is introduced as the other non-specified
negotiation offers are removed from the analysis. To address this concern a Heckrnan
probit model is specified with the same variables for each of the rare event models.

The Heckrnan probit models included the significant explanatory variables, from
the other rare events models with dependent variables omitted from the analysis, into the
selection equation. For example if win ratio is a significant explanatory variable in
category 1 rare events logit, the win ratio variable would be specified in a Heckrnan
probit selection equation for tests of category -1 independent variables. The presented

results are those explanatory variables that were statistically significant at conventional

levels in both the rare events logit model as well as the Heckrnan probit model. 207

 

20" The initiating states are Germany, Japan, Iraq, North Korea-China, North Vietnam.

2‘" The rare events logit is a robust analysis clustering on the referenced state. The presented rare events
logit models were also specified as logit models clustering on state, fixed effects logit models identified by
state. The same specification was significant across all of these different methodological specifications.
Additionally after removing influential observations (hat > .1) the results remain robust.

209

Table 3: Rare Events Logit Initiators Only

Log time
since
negotiations

Log of
negotiation
offers

Other
negotiation
discount

Lag total
defeats
(tens)

Cap ratio
change

Constant

Model 1
Outcome
-1
-.983"*
(.197)
.335
(1.022)
1929”"
(.71 1)
233“"

(.711)

-2.72*
(1.707)
N=306

Model 2
Outcome
-1
-l.53*"
(.137)
4.642"
(.796)
1.814‘

(1.163)

-9.112""'

(3.603)
7.533m
(2.622)
N=313

 

All results are one tai

 

ed tests, p < .075=* p<.05=**, p<.01=***

The cumulative number of defeats and the change in capabilities ratio over the

course of the war suggests that as the initiating is defeated more often and increasingly

unfavorable balance of power they will make sub-status quo negotiation offers.208 Upon

being disabused of this favorable opinion by both cumulative defeats and a deteriorating

balance of capabilities, the initiating states will begin to make comparatively unfavorable

bargains.

 

20’ However, defending states also make a calculation that conceding the issue in question is less
advantageous than would be a decision to resist aggression. Again there should be an expectation that
these states hope to gain more through fighting than they would from conceding the issue-goods in

questions.

210

These results appear to lend additional empirical merit to the ongoing bargaining

model of war that as the actual balance of capabilities and resolve is evidenced through

fighting this leads to a resolution of the conflict. In order to establish a contrast a similar

analysis was also performed upon all observations.

Table 4: Rare Events Logit Sub-Status Quo Offers All Cases

 

Model 1 Model 2
Outcome Outcome
-1 -1
Log time
since 57*" -.389"
negotiations
(.232) (.311)
Log
negotiation -.371 -. l 8
count
(.334) (.379)
Other
negotiation 2.8"" 2378*"
discount
(.407) (.724)
Cap ratio -2.45**
(1.117)
Win ratio 4.97"
change
(2.669)
Constant -l.671* -2.3 l "
(1.045) (1.21)
N=629 N=610

 

 

All results are one tailed tests, p < .075=* p<.05="'*, p<.01=***

The results indicate that theoretically similar variables appear to be significant in

the inclusive sample as opposed to only evaluating initiators only. The lower a state’s

relative material capabilities the more likely the initiating state is to make a lowered

211

issue-goods position offer. As a state’s win percentage improves from the previous
month it is less likely to make an ante bellum sub-status quo offer.

Whereas the initiators facing greater numbers of defeats and have seen the relative
balance of capabilities shift against them throughout the course of a war, appear to be
more likely to make category -1 offers. These findings correspond to a previously
elevated set of expectations being diminished by reduced capabilities and defeats at the
hands of their targeted opponent.

The broader sample instead stresses a dramatic change in the overall percentage
of wins that a state has enjoyed markedly decreases in the previous month. And it is the
raw level of capabilities rather than the relative decline in the relative balance of
capabilities that increases the likelihood of a sub-status quo bargaining offer. Regardless
both sets of variables lend credence to hypotheses three and four. Using these findings to
reduce the number of explanatory variables specified, a multinomial logit test for just the
autocratic cases is performed.209 The use of this sub-sample permits the introduction of

the political pressure variable into an analysis.

 

20” With respect to the control variables such as time since last negotiation offer and number of offers made,
these variables are retained across specifications in order to ensure that autocorrelation does not
compromise the presented results. Consequently in some specifications the log or square of time since
negotiation offer will appear or the log of actual count of negotiation offers will appear contingent upon
which variables permit conformity with regression assumptions.

212

Table 5: Multinomial Logit Autocratic Cases

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Model 1 Model 2 Model 3
Outcrl)me - Outcome 1 Outcirme - Outcome 1 Outcclrme — Outcome 1
Pseudo R _ Pseudo R2 _ Pseudo R2 _
=0.5350 “-503 =0.4999 “—493 =0.5226 N493
Log time
““9". -.808*" -.8“"" -.816*** -.973*** -.808*** -1.017***
negotiation
(.181) (.346) (.291) (.291) (.31) (.281)
Log
negotiation -.677*** .057 -.703*** .059 -.656*** .075
count
(.185) (.099) (.369) (.095) (.222) (.101)
Other
negotiation 196*" 3572*" 1.101 3003*" 2223*" 3359*"
discounted
(.623) (1.077) (.775) (.981) (.761) (1.023)
Log political
pressure 14.553" -1.352 9.748*** -1.052 16.005" -l.295
count
(9.773) (.619) (3.348) (.571) (7.03) (.677)
a“ “Y” 6448*" L685" 5.773'm .703 8675*" 1.346M
lll'athl‘l
(2.415) (1.097) (1.804) (.824) (3.264) (.748)
Win ratio a...
change -37.123 * 10.614
(12.408) (6.485)
Autocracy _ _ u
Polity Score 1'39 1'39
(1.348) (.365)
Win ratio -28.276 6.604
(21.505) (2.92)
causfl'ty -.024 -.01 1
ratio
(.012) (.006)
capgzi‘gt‘es -7.794*** -2.709*** -19.725*** -.58 -26.283*** -.901
(1.849) (.809) (6.902) (.663) (9.687) (.783)
““5“” -33.265*** -1 1.224" -32.662"* -2.053 -51.584*" -3.262**
(13.732) (5.186) (9.407) (1.65) (21.008) (1.607)

 

 

 

 

All results are one tailed tests, p < .075=* p<.05=**, p<,01=***

213

The dependent variables include only category 1 and -1 offers referenced against
no negotiation offers being made in that observation month.210 Of the hypotheses the
autocratic political pressure variable appears to have a significant impact upon the
likelihood of -1 category offers.211 Other independent variables that were found to be
statistically significant at conventional levels included the capabilities ratio between the
combatant states as well as the log of war duration.

Substantially war duration has an intuitive effect upon the likelihood of a sub-
status quo offers and status quo offers in some specifications. The greater duration of a
war is also joined by the balance in material capabilities in helping to define the
likelihood of less favorable negotiation offers. Finally if in the last month an autocracy
suffers a number of military defeats that reduce their overall percentage of wins and if a
greater number of political pressure incidents have occurred autocracies are more likely
to make sub-status quo offers. By contrast comparatively few of the variables specified
were statistically differentiable from zero in terms of their relationship with the likelihood
that a state would make a status quo negotiation offer.

The above hypotheses were tested in rare events logit models (King and Zheng
2001abc).212 The rare events logit model evaluates dichotomous dependent variables,

accordingly negotiation offers were evaluated separately (categories -1, 1), against no

 

2’0 There are no observations for a -1 offer being made by a democratic case. As such the elimination of
category 2 constitutes the most conciliatory offer that a democratic country will make. The autocratic cases
are compared to both status quo offers as well as -1 offers. Additionally, category 2 hypotheses tests do not
reach levels of statistical significance. The multinomial logit models were also tested for violations of the
IIA assumption and the appropriateness of the division of the observations into the dependent variable
categories. The models presented did not encounter these problems.

2” Autocratic cases also were specified for a control of the level of repression that a state was thought to
posses, this factor is controlled for by introducing J aggers and Gurr’s autocracy variable from the Polity III
dataset (1992).

2'2 King and Zheng (2001) illustrate the problems with the comparative rarity of success in a sample
coupled with finite sample size. In as much as the actual proportion of the population that consist of
negotiation offer-months is unknown, no specific correction for population frequency was specified.

214

negotiation offers (category 0). As with the above, a Heckrnan probit selection model is

also used.2’3

The presented results are those explanatory variables that were statistically
significant at conventional levels in the rare events logit model as well as the Heckrnan
probit model along with other diagnostic criteria.214

In order to further evaluate the hypotheses stated above another battery of tests is
applied to the autocratic observations, model 1 contains all observations, using a binary
analysis. Rare events logit models are specified for both category 1 and -l negotiation
offers referenced against no offers made that month for all autocratic observations.

The presented models also include an interaction term of capabilities ratio and
win ratio. The intention being to see if a state’s understanding as to its relative military
capabilities is adjusted by the outcome of battles rather than the two variables acting
independently.215 However, the interaction term does not enjoy statistical significance at
conventional levels for all observations or for the autocratic and democratic observations

separately.216

 

2'3 The Heckrnan probit models specified the significant independent variables from the other rare events
models with dependent variables omitted from the analysis, into the selection equation. For example if win
ratio is a significant explanatory variable in category 2 rare events logit, the win ratio variable would be
specified in a Heckman probit selection equation for tests of category 1 independent variables.

2 ’ The rare events logit is a robust analysis clustering on the referenced state. The presented rare events
logit models were also specified as logit models clustering on state and fixed effects logit models identified
by state. The same specification was significant across all of these different methodological specifications.
Additionally after removing influential observations (hat > .1) the results remain robust.

2" An interaction term using the difference in capabilities ratio and win ratio was also employed with
similar results.

2’6 The interaction effect variable's significance is more difficult to interpret, in as much as both the sign
and the z-score of the variable are subject to change conditioned upon the value of the cap ratio-win ratio
variable (Ai and Norton 2003). These findings call into question the assumption of the principal of
convergence, at least for status quo offers, that the previous expectations of fighting (capabilities) are
altered through battlefield success (win ratio). The relative unirnportance of the cap ratio-win ratio
interaction term conformed to the generally inconsistent or insignificant findings in the autocratic category
1 models.

215

Table 6: Rare Events Logit Status Quo Offers

Log time
since
negotiation

Log
negotiation
count

Other
negotiation
discounted

Log political
pressure
count

Cap ratio
‘win ratio

Win ratio
difference

Win ratio

Causality
ratio

Capabilities
ratio

Constant

 

 

Model 1 Model 2 Model 3 Model 4
Outcome 1 Outcome 1 Outcome 1 Outcome 1
N=708 N=391 N=392 N=388
-.564" -.525** -.527** -.538"
(.251) (.248) (.256) (.233)
904*" .096 .085 .068
(.264) (.084) (.072) (.074)
3461*" 2985*" 2986"" 3023*"
(.557) (.864) (.887) (.779)

-.366 -.406 -.629
(.587) (.618) (.725)
-.105*"
(.043)
6.351
(8.146)
2.744 1.947
(1.377) (1.831)
-.174
(.327)
.033 -l.539* -1.204 4.499"
(.012) (.989) (.848) (.879)
-3.907"* -1.727 -1.204 -.31 I"
(.701) (1.265) (1.074) (1.681)

 

All results are one tailed tests, p < .075="' p<.05=", p<,01=*u

Similarly the casualty ratio is generally insignificant, offering credence to the

216

possibility that autocratic states appear to be less averse to human losses than are their
democratic counterparts. The ratio of capabilities as well as change in win percentage
also appear to enjoy some significance at conventional levels, but these results only

partially conform to the rare events logit model using all observations.

Table 7: Rare Events Logit Category Sub-Status Quo Offers Autocratic Cases

 

Model 5 Model 6
Outcome -1 Outcome -1
N=442 N=443
L°g “"3“ time 2164‘" -.177
negotlatlon
(.766) (.619)
Time negotiation 065""
(.017)
Log negotlatlon __1 13 4.07]
count
(.153) (1.131)
Othjr negotiation 1629*" 2.65 5*"
lscounted
(.763) (.656)
Log polltlcal .987" 5. 6 6*"
pressure events
(.544) (1.131)
Win ratio -7.883"*
(2.46)
Capabilities ratio 43.221"
(7.95)
Capabllltles ratlo _ 6.82 6"
change
(4.171)
Constant -5.987* 7776*"
(3.936) (2.859)

 

 

All results are one tailed tests, p < .075=" p<.05=”, p<.01=***

Given the relative paucity of hypotheses tests with respect to status quo
negotiation offers, likely the more important sub-status quo offers represent incidences of
new battlefield information revising previous expectations. From the above results it
appears that most of the hypotheses tested do not enjoy statistical significance at
conventional levels. Instead category -1 offers should be more strongly identified by
states reassessing their issue-goods position on the basis of adverse battlefield events. A

rare events logit model is then specified for sub-status quo negotiation offers.

217

The relative share of battles that an autocratic state has won decreases the
likelihood that a sub-status quo negotiation offer will be made. Of the sampled autocratic
states with a comparatively unfavorable balance in material capabilities is more likely to
make a sub-status quo bargaining offer. A weaker relationship between change in
capabilities over the course of a war and the likelihood of a category -1 offer being made
is also found. Finally the autocratic political pressure variable also has a significant
effect upon the likelihood of a sub-status quo negotiation position.

Of the above hypotheses three, four, and six appear to have some of the greatest
impact upon the likelihood of an autocratic state making more effacing bargaining offers.
This provides more empirical evidence as to the importance of battles in refashioning
expectations as to the likely outcome of a war and consequently inspiring states to codify
their revised expectations with a negotiated settlement.

In the autocratic cases used for analysis it appears that status quo offers are not
considered to be especially onerous. This might be related to the general asymmetry in
power between autocratic combatants and their democratic antagonists, with the
autocracies being heavily overmatched in most observations. Consequently drawing a tie
in these wars might be considered a comparatively favorable outcome.

Similarly the relative level of material capabilities between states also has a
meaningful impact upon the likelihood that an autocratic state will be compelled to
relinquish more issue-good benefits in order to obtain a peace. The change in the balance
of capabilities between the beginning of the war is also differentiable from zero with

respect to the likelihood of a -1 peace offer. But if battles alter previous expectations as

218

to the actual balance of capabilities and therefore the likelihood of achieving victory over
the issue-goods at stake, this is not evident in the interaction term.

Amongst the stronger empirical findings is that in both the rare events logit
specifications and the multinomial logit models it appears as if political pressure is
significantly different from zero only in relationship to sub-status quo bargaining offers.
What remains to be determined is if the comparative weight that political pressure has
relative to other explanatory variables is equally important.

An analysis of how battlefield outcomes embodied by a state’s winning
percentage and the comparative balance in capabilities weight against the impact of
domestic political pressure is the next portion of my analysis. The predicted probability
of a sampled autocratic state making a sub-status quo offer is shown in the table below.
The values of the main independent variables are shifted from the 25th percentile to the

50th then to the 75'”. All other independent variables are held at their mean value.

Table 8: Change in Predicted Probability of a Category Sub-Status Quo Offer

 

 

 

Predicted Model 5 Model 5 Model 5
Probability of sub- Autocratic Change in Win ratio
status quo offer political pressure capabilities ratio

25* Percentile 0.015 0.284 0.020
50‘h Percentile 0.072 0.257 0.008
Difference 0.057 -0.027 -0.012
50m Percentile 0.072 0.257 0.008
75th Percentile 0.258 0.294 0.002
Difference 0. 1 86 0.037 -0.006

Change in the relative balance of capabilities substantially contributes to an

understanding of why less liberal states make effacing offers to their opponents. But as

219

the values of the intra-war change in material capabilities rises, the substantial effect
upon the probability of category -1 offer does not dramatically decrease. In fact the
predicted effect is the counter-intuitive increase in the likelihood of a sub-status quo
offer. The percentage of a state’s wins upon the likelihood of a sub-status quo bargaining
offer are consistent, but the substantial effects of this variable are muted when compared
to autocratic political pressure.

Of the independent variables correlated with the decision by an autocracy to
concede more in a negotiated settlement political pressure is amongst the more
statistically significant and stronger in its effects. An autocratic executive facing
mounting pressure might attempt to reduce the costs to domestic political actors by
altering the perceived amount of time before a war concludes or settling for a peace offer
of less utility but with a higher probability of being accepted?”

Whereas status quo offers made by autocracies are associated with variables
indicating both positive and negative prospects for a state’s military success, political
pressure does not have a strong effect upon the likelihood of a category 1 offers.
Category -1 offers are associated with a lower relative balance of capabilities and a
decrease in winning percentage in the previous month. Autocratic political pressure
enjoys a significant relationship upon the likelihood of a sub-status quo bargaining offer,
a finding that holds in both multinomial logit and rare event logit models as well as
across specifications.

The sample of democratic cases is substantially different in that there are no

clearly sub-status quo offers made by democracies in the sample of observations. Instead

 

2”Taking the simple mean of the time before war termination the typical category 1 offer takes place nearly
two years before the conclusion whereas the -1 offers are made less than two months before the termination
of a war.

220

status quo offers are the least beneficial offers made. Unlike the autocratic cases, the
explanatory variables that had a significant relationship with category 1 offers are
theorized to negatively impact a state’s chances for military success. This finding was
demonstrated in both multinomial logit models as well as the rare event model
specifications. Proceeding with the prior that the democratic cases appear to View status
quo offers unfavorably, a number of rare events logit models were conducted.218

The total number of unexpected losses, unexpected losses in the previous month,
and total losses were found to be a highly significant in terms of the likelihood of a state
making a category 1 offer. A democracy enjoying a higher percentage of wins amongst
total battles win ratio is less likely to make status quo offers. Although not a robust
finding, the casualty ratio variable does support the general finding that negative events
are associated with status quo negotiation offers.219

The United States makes up the overwhelming number of observations and enjoys
vast material superiority over their opponents. Given that expectations for military
victory should be comparatively high and that a mobilized democracy should be
successful in combat (Reiter and Stam 2002), battlefield losses should be disconcerting
outcomes. Total defeats, unexpected defeats, and a comparatively low win ratio foster a

comparatively negative assessment of a state’s prospects for attaining the issue-goods at

stake in the war and engender status quo negotiation offers.

 

2’8 The presented results also were those with coefficient values significant in both the Heckman probit

models as well as the rare events logit tests. The rare events logit is a robust analysis clustering on the
referenced state. The presented rare events logit models were also specified as logit models clustering on
state and fixed effects logit models identified by state. The same specification was significant across all of
these different methodological specifications. Additionally after removing influential observations (hat >
.1) the results remain robust.

2" As with the rare events logit models above, another test was conducted specifying the same models into
a Heckrnan probit analysis. The results echo the significance of cumulative battles, winning percentage and
recent unexpected losses in encouraging status quo negotiation offers.

221

Table 9: Rare Events Logit Status Quo Offers Democratic Cases

Log time
since
negotiation

Negotiation
count

Other
negotiation

Lag total
unexpected
defeats

Total
defeats

Casualty
ratio

Lag
unexpected
losses

Log
discounted
public
opinion

Win ratio

Capabilities
ratio

Constant

 

Model 1 Model 2 Model 3
N=152 N=151 N=151
919“" .843* 1515“"
(.257) (.53) (.392)

-.105 221"" 682‘"
(.163) (.182) (.185)
4534*" 5.048‘" 6.081 "*
(1.578) (1.708) (1.853)

.763**‘
(034)
.04***
(.006)
12295" * -3 1.906
(5.231) (24.638)
.52### .37**#
(.062) (.067)
-1.57*" -1.547*" 4299*"
(.595) (.438) (.547)
-3.778**
(2.605)
.033 -.018
(.022) (.037)
8785"" -1.825*" 5681“"
(l .063) (.646) (2.584)

 

 

All results are one tailed tests, p < .075=* p<.05=", p<_01=***

222

With respect to political pressure it appears that as an executive’s position is more
secure the less likely that the state will make status quo offers.220 To illustrate the
substantial effects democratic political pressure the discounted presidential approval
variable is shifted from the 25‘“ percentile to the 50‘“ then to the 75’“ percentile.
Comparatively secure democratic executives are less likely to make category 1 offers.221

The influence of political pressure upon the likelihood of a reduced negotiation
position demands a split in the cases between autocratic and democratic cases. After
dropping the category 2 offer negotiations the share of negotiation offers equals about 20
percent of the sampled observations. As such the greater uncertainty regarding an
executive’s future political position results in an increased likelihood of making ante
bellum status quo negotiation offers when other independent variables are held at their

mean value.

Table 10: Change in Predicted Probability of Status Quo Offers

 

 

 

Predicted probability of status quo Model 1
offers Democratic political pressure
25'“ Percentile .194
50‘“ Percentile .134
Difference -.06
50‘“ Percentile .134
75‘“ Percentile .065
Difference -.069

 

 

22° As noted above the democratic cases do not have a coded -1 offer and the category 2 offers are dropped
fiom the analysis. The presented rare events logit models were also specified as logit models clustering on
state, fixed effects logit models identified by state. The same specification was significant across all of
these different methodological specifications. The residuals from the robust logit analysis were regressed
on the previous four residuals, no statistically significant correlation was found at the 5% level in a two-
tailed test. Additionally after removing influential observations (hat>.l) the results remain robust.

22’ A comparison to other significant explanatory variables was attempted, rare events logit program failed
to converge.

223

Comparatively few of the independent variables are significant in their association
with less advantageous bargaining offers. Of those which are significant and presented
above the importance domestic political pressure appears to be substantially significant
even when other controls are specified. Given the vast numbers of models that can be
conducted with the above variables, the democratic and autocratic metrics for an
executive’s diminished political support is found to be amongst the strongest findings
across specifications and diagnostic tests. However, to demonstrate that these variables
are not artifacts of the presented results other measures of political pressure were also

employed.222

 

222 Several different manifestations of the political pressure variables included the specification of the time
and log of time until the election, the change in job approval between surveys, and the raw percentage
support for the executive. Of these the log of time until the election had a positive effect upon the
likelihood of a status quo offer being made. The log of time until an election was not robustly significant,
but was typically differentiable from zero when specified with the executive’s job approval numbers.
These significant findings are only definitive in the rare event logit models using similar specifications to
the presented results, not necessarily the other diagnostic criterion. The more robustly significant level of
executive job approval presents some credence to the political pressure variable without strict reliance upon
the discounted value of executive approval.

In the autocratic cases, a wider array of political pressure variables were employed, of these a number had
statistical significance at conventional levels. These significant findings are only definitive in the rare
event logit models using similar specifications to the presented results, not necessarily the other diagnostic
criterion. The onset of political pressure in an autocracy, indicating that a state has experienced a major
political disruption is correlated with a sub-status quo bargaining offer. The unmodified version of
pressure event counts and a dichotomous measure of whether or not a pressure event occurred both are
routinely significant in discerning the likelihood of a sub-status quo negotiation offer.

By contrast the logged time since a pressure event had minimal effect upon the likelihood of a category -1
negotiation offer, seemingly offering validation of the assumption that elapsed time since a pressure event
is not a significant factor in a dictator’s calculations. However, a count of pressure events then divided by
the number of months since the last pressure event did enjoy a statistical significance in some
specifications.

The most statistically substantial results are illustrated in the above regression tables. Theoretically the
time discounted value of a democratic executive’s job approval and the growing danger of deselection
evidenced by an increasing number of pressure events for an autocrat are the most satisfying variables
encapsulating political pressure.

224

Table 11: Definitions of Changes in Negotiation Offers

 

Upward Move Downward Move
Sub-status quo to no offer (-1 —> 0) No offer sub-status quo (0 —r -1)
Status quo to no offer (1 —-) 0) No offer to status quo(0 —> 1)

No offer to above status quo (0 —r 2)

An alternative dependent variable was also specified to evaluate if the political
pressure variables remained significant. With the insight provided by the above results,
an inference of the relative superiority of different kinds of bargaining offers and no
offers being made was constructed for both democratic and autocratic cases. The new
dependent variable uses the same negotiation offers as the above analysis, but examines
the change in negotiation offers rather than the bargaining offers themselves. 1 The change
in negotiation position variable assumes three values, a less favorable negotiation
position, no change, and a more favorable negotiation position.

When a state moves from no negotiation offer to a status quo offer this is coded as
a downward move on the scale. A sub-status quo offer is also recorded as a downward
movement in negotiation offers. Moving from either a sub-status quo offer or a status
quo offer to no negotiation is recorded as an upward movement. A negotiation offer for
greater benefits than at the onset of a war is considered an upward negotiation offer, but a
change to no negotiation offer is not recorded as a downward move.

Only two models containing both a political pressure variable and other
theoretically salient independent variable were found to be statistically significant in rare

223

event logit models and the Heckman selection specifications. The autocratic cases

 

223 As with the different negotiation offers variables the change in negotiation position was first evaluated
in both an ordered and stereotype logit models with consistent violations of their foundational assumptions.

225

appear more willing to move to a favorable bargaining position as winning percentage
increased and change in capabilities ratio improved. Higher values in the autocratic

political pressure variable were associated with a decreased likelihood of making a more

favorable bargaining offer.

Table 12: Rare Event Logit Upward/Downward Movement

Upward Downward
Autocracies Democracies

 

Log time
since -.835 .628
negotiations
(.587) (.561)
Log
negotiation .668‘" .245
count
(.111) (.448)
Lag
unexpected .238
losses
(194)
Cap ratio .83 6"
change
(.369)
Lag total
unexpected 346*"
losses
(.052)
Log
Pressure -.417**
count
(.24)
Log Public
opinion -.555***
discount
(.058)
Win ratio 1.727"
(.819)
Constant -3.197*" -5.756*"
(1.614) (.058)
N=482 N=l45

 

 

 

All results are one tailed tests, p < .075=* p<,05=", p<,01=u¢

 

The percentage of downward negotiation offers was .051 of the observations and upward negotiation offers
constituted .046 of observations

226

The only democratic specification found to be significant was in the downward
movement in negotiation position. Total unexpected losses is associated with an
increased likelihood of a downward movement in negotiation position, the number of
unexpected losses in the previous month is signed in the same direction but is not
statistically significant. The more secure an executive is in terms of the level of political

pressure the less likely that a downward movement in negotiation position will be made.

Table 13: Change in Predicted Probability of Movement in Negotiations

 

 

 

Downward move Upward move
Predicted probability Democratic political autocratic political
pressure pressure
25‘“ Percentile .025 .026
50‘" Percentile .021 .022
Difference -.004 -.004
50‘" Percentile .021 .022
75‘h Percentile .015 .019
Difference -.006 -.003

 

Given the overall paucity of significant results in both the rare event logit models
and the heckman selection models, it would be hazardous to extrapolate a great deal from
the findings. Generally the political pressure variables performed as well as any other

explanatory variable in these models and conformed to the results from the negotiation

offer models.

227

Conclusion

Negotiation offers ~are used in my analysis as an empirical demonstration of a
state’s changing appreciation for the likely issue-goods benefits it can achieve. An
alternative assumption holds that objective criteria are assessed that determine the actual
balance of resolve and capabilities, allowing states to reevaluate previous expectations as
to the likely outcome of a war. Referencing this principal of convergence literature, a
number of hypotheses were evaluated that would indicate a state had changed their
expectations.

What is striking about the results is both the relative significance and substantial
importance of the political pressure variables. The comparative level of autocracy, war
duration, military expenditures, and casualties are generally un-differentiable from zero
in defining the likelihood of negotiation offers. Variables associated with an explanation
as to why wars terminate are largely mute in the sample examined. This would also
conform to the case studies examined in the previous chapters. The importance of
anticipated punishment, and war selection assessed in a post-hoe fashion were not found
to be compelling reasons for changes in issue-goods position.

My theory is also offered as an alternative to the inferred unitary actor
assumption. If states are functionally rational actors Operating at the behest of the
executive or an aggregated assessment made by governmental elites, this should lead to
an objective assessment of the likely outcome of a war. Eventual war outcome is
understood as the relative resolve and capabilities of the opponents fighting for the issue-
goods at stake in the conflict. However the objective factors that are thought to dictate

the comparative resolve and capabilities of the opponents are secondary in importance to

228

the political pressure variables. An executive threatened by deselection or reduced
political authority should be more likely to make comparatively unfavorable negotiation
offers.

The typical decline in rallies of public support for executive following the onset
of hostilities means that a leader faces a more skeptical public in the future. After an
initial upward movement in support the electorate returns to its more normal state defined
by demographic factors in an assessment of costs and benefits (Gelpi and F eaver 2004,
149-176). As the costs of a war increase relative to benefits this leads to a retraction in
support for mobilization that can be addressed through an alteration in war policies.

As the public’s appraisal of a president or prime minister is comparatively high
and an election is irnrrrinent this should reduce the imperative to alter war policies away
from what an executive prefers. By contrast when an executive’s popularity is high and
an election is further off or when their popularity is low this increases the likelihood that
an elected leader will make comparatively unfavorable negotiation offers. Notably this is
found to be the case in the Vietnam War where the preferred issue-goods of Nixon were
modified in order to secure a settlement and prevent deselection.

For the autocratic cases are examined, no clear association between negative
battlefield events and status quo offers is found. Instead sub-status quo offers are
consistently associated with negative events. The lower an autocratic state’s capabilities
relative to its opponent and the lower its win ratio the more likely that the state will make
sub-status quo offers. The autocratic political pressure variable also enjoys statistical
significance in increasing the likelihood of category -1 negotiation offers. In both the

case of Nazi Germany and Imperial Japan the increased difficulty in scoring military

229

victories and the steady erosion of their relative capabilities led to sub-status quo ante
bellum offers.

A challenge for these findings is that sub-status quo offers are sometimes
tantamount to the regime’s destruction, so why would an executive seek to accommodate
domestic political pressure just to surrender his/her position? This is partially addressed
by the model of negotiation movement rather than negotiation offer. A tentative finding
of an increased incidence of elite dissent decreases the likelihood of an “upward”
movement in an autocratic state’s negotiation position.

Another factor true of both democratic and autocratic cases is that the dependent
variable is by necessity broad and somewhat clumsy. The lack of nuance in these
variables means that the comparatively favorable or unfavorable movements within a
category are not detected. As indicated in the case studies chapters, examples of sub-
status quo offers might reflect an executive’s willingness to accommodate domestic
political pressure while maintaining his political position.

Frequently it is found to be the case that an autocratic executive is the one more
reticent to initiate negotiations and to make the necessary concessions in order to
facilitate a peace settlement. Demonstrated by Mao being compelled to open
negotiations and then eventually acquiesce to a revised settlement in part due to political
pressure. Similarly Khomeini was forced away from his long-standing goal of destroying
the Iraqi regime. Even in these cases coded as status quo settlements, political pressure is
a significant causal factor.

As fighting reveals the relative capabilities and resolve of combatants this is

theorized to eventually end in mutual acceptance of this balance and an agreement

230

reflecting this understanding. The above results indicate a potential problem with the
assumption of an effectively unified cost-benefit assessment conducted by a state. At
least one of the antagonists must reduce their issue-goods position in order for a
settlement to be reached.

However, the less advantageous bargaining offers are not explicated by just a
reference to casualties, negative battlefield events, or the distribution of capabilities.
Instead as an executive’s constituents appreciate the current utility of a war less favorably
and apply pressure for changes in war policy, this is sometimes accommodated by a
change in issue-goods negotiation position. The ability for domestic actors to assess a
war’s utility and instigate changes is a challenge to the empirical saliences of the ongoing
bargaining model of war.

Whereas the selection of wars can have a profound impact upon the likely
outcome of the conflict, the implication that the war is only assessed after the result is
known is not convincing. Instead the way the war is fought, the resources mobilized, and
ultimately the issue-goods settlement are all in part informed by political pressure
compelling changes in executive policy.

The empirical study above takes explanatory variables of superior validity in
order to explain the likelihood of a change in negotiation position. The findings indicate
that political pressure is one of the principal factors driving less advantageous bargaining

officers in the quantitative models.

231

Chapter 5: Changes in Strategy

Whereas issue-goods negotiation position dictates for what a state is fighting,
military strategy is the manner in which a state fights. Alteration of strategy may help
change perceptions of an executive’s constituency as to the utility of a war. Instead of
forgoing an executive’s preferred issue-goods or increasing mobilized resources, a
change in military tactics provides an appealing way for an executive to mitigate political
pressure.

A number of hypotheses as to why strategy change occurs have been presented.
The influence of changes in key variables such as casualties or body counts, the lack of
material resources, and the desperation implied by a leader facing punishment. In the
case study chapters an alternative dynamic is presented, where strategy is changed
contingent upon the influence of elites lobbying for adaptation in tactics and an effort by
the executive to alter the cost-benefit assessments of domestic political actors.

As with the issue-goods analysis, a quantitative test as to the reasons for strategy
change is conducted. The task is again to test alternative hypotheses against the political
pressure theory in order to see which factors are the most important in determining
strategy change. Additionally, the relative weight that different variables exert upon the
likelihood of strategy change is also assessed.

Substantively strategy represents one of the major arbiters of military victory and
defeat (Starn 1996; Malkasian 2002; Biddle 2004). The study of strategy has largely
been limited to the comparative efficacy of different strategies. For the most part change

in strategy has been limited to qualitative studies mainly using democratic great powers

232

(Gartner 1997; Avant 1993). The following chapter constitutes one of the first efforts at a
quantitative anlaysis of strategy change across cases.

Returning to the narrative of the political pressure theory, the executive’s sense of
endangerment can induce an alteration away from the leader’s preferred war policies.
Whether at the behest of a lobbying effort or driven by the need to assuage comparatively
negative assessments of the war, changes in strategy can substantially change the
outcome of a conflict.

The possibility of an intra-war assessment by an executive’s principals means that
not only are assessments of a war not suspended until the final outcome is determined,
but that the outcome is in effect changed because of political pressure. If an executive is
wedded to certain issue-goods positions and military tactics designed to achieve these
aims, being compelled to alter war policy might result in an improved strategy being

employed.

Strategy Change

Three broad categories are used to define strategy; attrition, maneuver, and
punishment (Stam 1996; Reiter and Meek 1999). Strategy is defined as maneuver if a
state is attempting to utilize mobile warfare and encirclement operations (Stam 1996).
The Blitzkrieg implemented by Nazi Germany is a ready example of a maneuver strategy.
Attrition is an attempt to destroy an opponent in a series of pitched battles or through
exhaustion, World War I’s trench warfare serves as the archetype of this strategy.
Punishment is defined as an attempt to inflict pain upon an opponent. Guerrilla attacks,

blockades, and strategic bombing are considered manifestations of this strategy.

233

A hierarchy is used by Stam to define the relative effectiveness of a strategy
contingent upon what strategy an opponent plays (1996). A state can play three strategies
and so can an opponent, for a total of nine potential interactions. The military strategy
hierarchy also implies whether a state is using the strategies offensively or defensively.

To convert this metric for the purposes of my study, the state that appeared to
enjoy the strategic initiative during that particular calendar month was determined.224 To
illustrate in the Western Front of World War II, the Anglo-American forces are
considered to be on the offensive from the Normandy landing until the Battle of the
Bulge (Ardennes Offensive) in December of 1944. For a brief time the Germans returned
to the offensive, only to again lose the initiative to the Allies in the following month.
Likewise, US ground forces were mostly in a defensive posture throughout the early
months of the ground conflict in Vietnam, only to take the initiative in 1965 against their
communist opponents.

The three strategy taxonomy has been the subject of criticism. Biddle has
described the difference between attrition and maneuver strategies to be somewhat
subjective, because a failed attempt at maneuver may be interpreted as being an attrition
strategy (2004). This challenge to the coding of strategy is addressed through the
creation of two variables strategy employed and strategy attempted.

The battle of Kursk on the Eastern Front of World War II is illustrative of the
distinction. The German command planned an encirclement of Soviet force by breaking
through Russian lines at key points with superior forces and then disrupting the rear

areas. The salient (bulge) at Kursk seemed a prime target for a resurrection of blitzkrieg

 

22" While the hierarchy presented by Starn (1996), Reiter and Meek (1999) was designed for the purpose as
such, it is still instructive with respect to which strategies are relatively superior given the opponent’s
strategy in the same period.

234

on the Eastern Front. However, the obvious German preparations and equally transparent
vulnerability of the Soviet forces in the pocket led to substantial defensive preparations.
The attack faltered in its ascribed goal of a maneuver strategy, it instead degenerated into
the great armored struggle of attrition for which this campaign is famed. This dyad-
month is coded for the Germans on the Eastern Front as an attempted maneuver strategy
and an actualized strategy of attrition.

An example of a strategy change that is not coded for either attempted or
actualized strategy change was that of the United States in 1968. With the conclusion of
Westrnoreland’s tenure in command of US forces strategy changed markedly under his
successor General Abrams. The new tactics moved away fi'om the large battlefield
formations attempting to force engagements with larger communist units, instead
operations attempted trying to control and pacify territory. While certainly a marked shift
in tactics, this is considered another form of attrition and is not coded as a strategy
change.225

The dependent variable in the analysis is attempted strategy change. Although
there is little difference between these two measures in terms of their frequency, the
reason for this is theoretical. Attempted strategy change is a superior choice in that
political pressure may encourage an alteration in tactics, but not necessarily reflect the
ability to execute a new strategy.

The dependent variable is coded as 1 if an attempted strategy change occurred in a

calendar month and 0 otherwise.226 For example the initiation of the Tet Offensive

 

22’ For a discussion on the various types of attrition strategies and the relative efficacy of these tactics see
Malkisan (2000).

22" To clarify, this is the month in which the strategy change was attempted not necessarily the month in
which the decision for a strategy change was made.

235

changed North Vietnam’s strategy from punishment to attrition, a l is reCorded for
February 1968. Hanoi then reverts back to a punishment strategy in the following month
resulting in a 1 being coded for March 1968 as well.

The political pressure hypothesis is challenged by simpler alternative
explanations. The first is trial and error, by which a comparatively inefficient strategy is
changed in order to improve a state’s military fortunes. Progressively inferior strategies
should be replaced over time with superior ones. An executive’s critical appraisal of key
indices such as casualties, enemy body counts, and losses of military assets are used as
the impetus behind strategy changes (Gartner 1997).

A change in strategy might also be employed in order to accelerate the conclusion
of a war (Pillar 1983). As military pressure is employed to increase costs and compel an
opponent to relinquish certain issue-goods in order to facilitate a conclusion to a conflict.
Another proposition is forwarded by Goemans that contends that wars are reviewed by
domestic interests in a state after the fact (2000). Anticipating the prospect of
punishment the executive will execute a more aggressive strategy, in terms of the
resources devoted to the conflict as well as the manner in which the war is conducted.

Military strategy can be altered for a number of reasons. I contend that strategy
change is not entirely a function of an executive’s unitary preferences or limited
capabilities to undertake alternative tactics. Instead deviations away from strategies of
greater personal efficacy to the executive are sometimes used as a means of mitigating

political pressure.

236

Hypotheses Tests

An obvious reason for an executive to change strategy is that a given strategy is
comparatively inefficient and inferior strategies will be replaced by superior ones. A
state’s strategy is ranked from 1 (best) through 9 (worst) on the basis of the antagonists
strategies in interaction (Starn 1996). A variable is formed by taking the rank of the
strategy attempted in the previous month. A less efficient strategy based upon the
previous state’s fighting should be altered in order to improve a state’s utility.

HI : A less eflicient strategy in the previous month increases the likelihood of an
attempted strategy change in the following month.

Changes in strategy by one’s enemy might either encourage a state to mimic the
other’s tactics or adjust to the new strategy. To capture the effect of states countering the
changes in the military tactics of their opponents’ the number of months since an
opponent’s change in strategy is summed, added to one, and then logged.

H2: If a state ’3 opponent has changed strategy this increases the likelihood of an
attempted strategy change in the following months.

Strategy choice might be restricted by a state’s material capabilities (Reiter and
Meekl999). Even if a state is able to change military strategy, it may be inadvisable to
do so as it invites a crushing defeat by the superior material capabilities of an opponent.
The COW’s capabilities index is utilized to create a capabilities ratio variable, with the
reference state’s material capabilities divided by its opponent’s (Small and Singer 1982).
Another variable is also created which illustrates the change in the balance in capabilities
over the course of the war. The difference in capabilities ratio variable takes the initial

capabilities ratio and divides this by the current capabilities ratio.

237

An alternative formulation of this hypothesis takes the COW’s military
expenditures variable and dividing this variable by the figure for military personnel. The
troop quality variable is designed to capture the level of training and equipment that
opposing forces should posses (Reiter and Meek 1999). The troop quality of the
reference state is then divided by their opponent’s value. A state with a better equipped
and trained force likely has a greater range of options in terms of strategy.

H3 : The greater the capabilities of a state relative to its opponent the more likely

that the state has the ability/willingness to alter strategy, increasing the likelihood

of an attempted strategy change.

Military setbacks might serve as the impetus for a change in strategy. A number
of different battles variables are brought to bear in order to examine this possibility. The
unexpected battles variables for the previous month and a second. variable totals these
events throughout the course of the war. Other battles variables include total defeats,
defeats in the previous month, win ratio, and the change in win ratio in the past month.
The more adverse a state’s military position in the aggregate or due to recent events the

more likely it will be to attempt to change its fortunes through new tactics.

H5 : Negative battlefield events should increase the likelihood of an attempted
strategy change.

As a state absorbs greater human costs, the more likely they are to change military
strategy in order to improve their prospects for battlefield success. This variable also
approximates the importance of key variables providing the impetus behind changes in
strategy (Gartner 1997). The number of fatalities that a state has suffered over the course
of the war is specified. The casualty ratio variable takes the referenced state’s previous
casualties and divides this by their opponent’s losses. This also captures the potential

importance of body counts serving as a metric of success (Gartner 1997; Slantchev

238

2003a). A state performing comparatively poorly in terms of personnel losses is more
likely to change strategy to improve battlefield outcomes.

H6: Higher casualties or casualty ratio increases the likelihood of an attempted
change in strategy.

As a war continues on without resolution a state might seek an alternative strategy
to hasten a settlement. The logged number of months a war has been ongoing is
specified. Another control was the cumulative number of battles that had been fought
until the previous month. An alteration of strategy might serve as a means of shortening
a war without necessarily surrendering issue-goods utility.

H7: The longer a war ’s duration the greater the likelihood of an attempted
change in strategz.

As an executive feels a dissipation of political authority or an enhanced likelihood
of deselection, this should increase the likelihood of an alteration in war policies to
improve the utility of his/her domestic political actors. The use of a different strategy
might improve the perceived likelihood of achieving the benefits in a conflict, reducing
the time to a settlement, and reducing the costs of continued fighting.

H8: Endangerment of an executive ’s position or political authority increases the
likelihood of an attempted change in strategy.

Analysis
A rare events logit model with robust standard errors was used in the following
specifications (King and Zheng 2001).227 The logged and squared number of months

since the last attempted strategy change was included in order to control for how long a

 

227The “actual” percentage of strategy change in the population is unknown, therefore the sample is not
corrected to balance for the population weight.

239

state has been executing a strategy.228 As was a count of the number of times that a state

has attempted to alter strategy throughout the course of a war.229

The presented results
include those variables that were significant in robust rare events logit specifications and
subsequent diagnostic tests.230

As the relative balance in casualties becomes less favorable for the democratic
state the more likely they are to attempt to change strategy. The only test of the battles
hypothesis that was significant was the number of defeats in the previous month, even
this variable was not significant in model five as was the change in capabilities ratio. The

political pressure variable evaluating is robustly significant across specifications.231

 

22" The square of time since strategy change was included in the presented specifications to better control
for a return to a previous strategy immediately following a strategy change. Also, the logged number of
revious attempted strategy changes was dropped for the better performing count.

29 The rare events logit is a robust analysis clustering on the referenced state. In order to insure that
autocorrelation is not a problem lagged error terms, 14 months back were placed into an OLS model and
regressed against the error term at time t. An additional control variable specifying the log of time since
the last negotiation offer was made by that state, this was significant in model 1 of the autocratic cases but
not significant in democratic specifications. Without having strong theoretical predictions as to what its
effect should be the variable did not appear in the reported results. An additional variable discounting
opponent negotiation offers was not significant in autocratic specifications and did not converge in
democratic models.
23° Some of the independent variables were significant when applied to all cases, and consequently in the
absence of either of the political pressure measures, however these results are not reported independently.
Diagnostic tests included a logit analysis with standard errors clustering upon the referenced state fi'om
which both the residual and leverage variables were obtained. The residuals were placed into an OLS
model with up to five lagged residual values, the presented results did not have correlation between the
residual variable and these lagged terms that were significant at the 5% level in a two-tailed test. In order
to alleviate problems arising fiom influential observations those variables that scored higher than .1 in the
‘hat’ statistic were then eliminated and a second rare events model was performed on the remaining
observations. Presented results are those models whose variables testing the above hypotheses retained
gignificance at conventional levels even after discarding influential observations.

3 Several different manifestations of the political pressure variables included the specification of the time
and log of time until the election, the change in job approval between surveys, and the raw percentage
support for the executive. In various models these alternative specification of the democratic political
pressure variable wee assessed to see if they affected the likelihood of an attempted strategy change. Of
these the log of time until an election was not robustly significant. The raw percentage of popular support
that an executive enjoyed was also statistically inert. Whereas the change in an executive’s approval rating
between surveys appears to be routinely significant. These significant findings are only definitive in the
rare event logit models using similar specifications to the presented results, not necessarily the other
diagnostic criterion. The more robustly significant change in an executive’s job approval presents some
credence to the political pressure variable without strict reliance upon the discounted value of executive
approval.

240

Table 14: Rare Events Logit Attempted Strategy Change Democratic Cases

Modell Mode12 Model 3 Model4 Model 5

 

Strategy change count .643" .608" .322 .623" .073
(.337) (.353) (.273) (.353) (.903)
Square of strategy duration .0001 .00003 .0003 .0002 .0004

(.0009) (.0009) (.001) (.0008) (.001)
Log of public approval discountedl-l.827*** -1.923"* 2124*" -1.774*"”" -2.205"""*
(.404) (.375) (.596) (.401) (.194)

 

Lag defeats .189**" .174
(.073) (.876)

Cap ratio difference 934*" .1 13
(.149) (1.825)

Causality ratio 6351"" 6767*"

(2.2) (2.449)
Constant 2.427 2279"" -3.049"* -2.922*" -1.985
(.452) (.4) (.359) (.5) (2.823)

 

N=154 N=154 N=154 N=154 N=154
All results are one tailed tests, p < .075=*, p<.05=", p<_01=***

Moving beyond statistical significance, the following table illustrates the
predicted probability that a state will attempt a new military strategy conditioned upon
changes in the values of the key explanatory variables. Predicted probabilities are created
by moving the independent variable from the 25‘“ to the 50‘“ and then to the 75‘“

percentile of their values.

 

In the autocratic cases, a wider array of political pressure variables were employed, of these a number had
statistical significance at conventional levels. These significant findings are only definitive in the rare
event logit models using similar specifications to the presented results, not necessarily the other diagnostic
criterion. The raw count of political pressure events is robustly significant at conventional levels in
increasing the likelihood of an attempted change in strategy. The time since a political pressure event
occurred has no meaningful impact upon the likelihood of a state attempting to alter military strategy. The
onset of political pressure in an autocracy, indicating that a state has experienced a major political
disruption is also correlated with an attempt to change military strategy. The unmodified version of
pressure event counts and a dichotomous measure of whether or not a pressure event occurred both are
routinely significant in discerning the likelihood of an attempted change in military strategy.

The most statistically substantial results are illustrated in the above regression tables. Theoretically the
time discounted value of a democratic executive’s job approval and the growing danger of deselection
evidenced by an increasing number of pressure events for an autocrat are the strongest variables.

241

Table 15: Predicted Probability of an Attempted Strategy Change

 

 

 

Predicted Model 4 Model 3 Model 3
probability of Democratic Model 4 Democratic Change in
attempted political Casualty ratio political capabilities
strategy chafle pressure pressure ratio
25"“ Percentile .022 .005 .017 .086
50th Percentile .013 .006 .009 .091
Difference -.009 .001 -.008 .005
50‘h Percentile .013 .006 .009 .091
75'“ Percentile .005 .007 .003 .106
Difference -.008 .001 -.006 .015

The substantial significance of the political pressure variable is relatively strong
given that the percentage of months observing an attempted strategy change is 4.33%.
Noting this, both political pressure and the change in capabilities ratio over the course of
a war have non-trivial effects upon the likelihood of attempted strategy change.

What is striking is that casualties are not especially important in determining the
likelihood of an attempted change in strategy. Capability ratio has a potent affect upon
the likelihood of an attempted change in strategy, but is only'intermittently significant
across specifications. The impetus behind strategy change does not appear to emerge
principally from the objective criterion of material resources or from the relative
efficiency in casualties instead it appears that political endangerment is likely the most
significant factor tested here.

The test of the autocratic political pressure hypothesis was less successful,
enjoying some statistical significance when specified with either the log of war duration
or the cumulative number of battles. Of the models listed in the table below only model 1

passed all diagnostic tests. The logged duration of a war and the number of cumulative

242

battles enjoyed significance in these models, but with the opposite sign than was

theorized.

Table 16: Rare Events Logit Attempted Strategy Change Autocratic Cases

Strategy change count

Square of strategy duration
Log number of pressure events
Log of war duration

Cap ratio change

Win ratio

Number of cumulative battles
(Tens)

Lag Stamrank

Constant

 

 

 

Model 1 Model 2 Model 3 Model 4
0663*" 0682"" 0689*" 0696*"
0.176 0.174 0.191 0.189
0.0005‘" 0.0005*** 0.000?" 0.0004“I
0.0002 0.0002 0.0002 0.0002
0.452" 0451*“ 0.409" 0.43"
0.199 0.178 0.191 0.216
-l.104*** -1.169*** -1.172*** -1.l33*”
0.218 0.27 0.266 0.317

-0.391 -0.23
0.292 0.518
-0.009
0.568 _
-0.06** -0.063** -0.052"'** -0.059"
0.029 0.029 0.02 0.033
0.043
0.11
-0.817"‘* -0.608 -0.367 -0.742
0.388 0.736 0.573 1.251
N=5 12 N=509 N=5 12 N=5 12

All results are one tailed tests, p < .075="' p<.05=**, p<,01=u*

The longer a war continues and the greater the number of battles fought,

autocratic states tend to retain their same strategies. The path dependency of this finding

is a little curious, but my favored interpretation is that an executive is not inclined to alter

strategy until acted upon by threats to his political position.

When comparing the changes in the predicted probability of an attempted change

in strategy it appears that political pressure is many times more important in its effects

upon the likelihood of an attempted strategy change. The employment of a

243

comparatively static strategy employment might be due to a lack of responsiveness to
battlefield outcomes, casualties, and capabilities. Whereas a democracy that is faced with

rising numbers of battle deaths has a stronger imperative to alter strategy.

Table 17: Changes in Predicted Probability of Attempt to Change Strategy

 

 

 

Predicted probability of Model I
attempted strategy Autocratic political Log xgdlln’ation Lag curirdlllgteilvb battles
change pressure
25“I Percentile .378 .025 .005
50‘“ Percentile .444 .013 .004
Difference .066 -.012 -.001
50'“ Percentile .444 .013 .004
75‘“ Percentile .512 .007 .003
Difference .068 -.006 -.001

An interpretation of the path dependence demonstrated in the sampled autocratic
cases would be that the executive of an autocracy tends to have an interest in maintaining
current war policies. Once having invested personal and political prestige into a war an
executive is more likely to continue doing so. Perhaps to avoid a demonstration of
weakness or admitting to incompetence, aversion to continued costs of a war become
apparent to the domestic political actors.

However autocracies appear to implement somewhat more effective strategies
than democracies of the sample with a mean score of 4. 16 as opposed to 6.85 (1=best,
9=worst). Suggesting that a change in strategy offers a lower amount of improvement in
strategy would be realized than would be true of democratic cases. The rate of attempted

strategy changes in autocracies is somewhat higher than in the more liberal states 6.91%

244

versus 4.33%. The reasons for military strategy may be somewhat more stochastic than
the reciprocal nature of negotiations.

The importance of relative casualties and the outcome of battles had no
substantial effect on the likelihood of a change in strategy. The relative balance in
capabilities or the change in this balance has a meaningful effect upon the likelihood of a
change in battlefield tactics. While on average the autocratic cases have better strategies
than did their democratic counterparts, the rate of attempted changes in strategy is higher
for the autocratic cases than is the case for the democratic observations.

The autocratic cases had no meaningful correlation between the level of autocracy
and the likelihood of change in military strategy. Additionally it appears that war
duration while significant, reduced the likelihood of a state attempting a change in
strategy. Both of these variables appear to discount the importance of a change in
strategy being driven by an intermediate level autocracy seeking to gamble for
resurrection at the end of a war.

The importance of the objective variables are of limited significance. Capabilities
ratio, battlefield deaths, the comparative efficacy of the current strategy, are all of limited
significance or muted empirical effect. The objective measures that are theorized to
influence the likelihood of a change in strategy do not appear to be especially
consequential in the sample of autocratic observations. Instead as with the democratic
cases, the threat of deselection appears to be the most cogent explanation as to what

drives attempts to alter a state’s strategy.

245

Conclusion

The previous chapter identifies the correlates of why states annunciate
comparatively unfavorable bargaining offers. The findings hold that while both
democratic and autocratic observations have alternative hypotheses with empirical
salience, it is the political pressure variables that appear to be the most substantial. The
same narrative appears to be at work in the above analysis.

A unitary assumption of a state assessing the relative balance of capabilities and
resolve demonstrated in the battlefield, does not appear to be well established in the
above analysis. While other hypotheses are found to be significant, especially in the
democratic cases, for the most part objective indicators of the efficacy of a strategy are
not as substantially important as is political pressure. Instead of objective metrics of a
state’s success in a conflict, differentiated interpretations of a war’s utility is instead
made.

This differentiated assessment is inferred in the institutional literature, as war
selection is the source of post-hoe review and sanction. The reasoning for this finding is
that an executive either derives more utility from a war fought for private goods or has
invested political clout into the effort and is recalcitrant to alter war policies. Regardless
it is intra-war assessment rather than a unitary appraisal or post-hoe review that appears
to be better supported by the above analysis.

As with changes in issue-goods benefits, an alteration in strategy appears to
follow indications that an executive’s political position is being jeopardized. In response

to an increased likelihood of deselection the prevailing tactics that a state is employing

246

will be altered in response to lobbying by domestic political actors or as a means of
altering the cost-benefit calculation of his/her principals.

An autocratic executive pursues strategies that are in line with his preferences. It
does not appear to be the case that these less-liberal states assess the merits of a strategy
on the basis of battlefield defeats or losses in soldiers. Likewise, the efficacy of a
strategy based upon the Stam hierarchy is insignificant in catalyzing changes in tactics.
These findings are reinforced by the autocratic case studies.

The North-first faction of the Hanoi Politburo believed that aggressive strategies
were undermining the regime’s standing and concurrently the populace’s willingness to
continue to fight. The Baathist elites also asked for a less aggressive tactics to insulate
the population from greater costs. Change in attempted strategy might also reflect a
gamble to end the war before domestic political pressure usurped the executive, such as
Khomeini’s offensives of 1986 or Hitler’s assault on the Allies through the Ardennes.
Either because of lobbying by elites or the need to alter the cost-benefit assessment of
domestic political actors, strategy is altered. The political pressure hypothesis tests are
comparatively strong when compared to alternative causes of an attempted change in
military strategy across the autocratic cases selected here.

Given that strategy is found to be comparatively static until acted upon by outside
influences. Even when a strategy is comparatively inefficient, this does little to increase
the likelihood of an alternation in the way a war is fought. Rather than a post-hoe review
of war policies, domestic political actors fearing punishment for themselves or loss of

their positions of influence are instead likely to attempt to alter the outcome of a conflict.

247

The substantial importance of this finding is that military strategy is considered one of the
most important arbiters of victory and defeat in war.

A larger portion of the hypotheses tests were significant at conventional levels in
the democratic cases. Concurrent with Gartner’s study of changes in strategy, as the
“body count” of the opponent is smaller relative to a referenced state’s own casualties
this increases the likelihood of a change in strategy (1997). The case studies bear out this
relationship. As the US increased its advantage in terms of capabilities relative to Japan
and the ratio of American battle deaths grew less favorable, a punishment strategy was
employed.

A favorable change in the ratio of capabilities can also help bring about a change
in strategy. This might be reflective of the United States’ ability to execute a wider
spectrum of strategies given its superiority in material resources. The United States had
the ability to transition to a maneuver strategy following the Normandy invasion,
something that Germany attempted but was unable to sustain in its assault on the
Ardennes.

Recent defeats also appeared to be differentiable from zero, although its
substantial affect remains undetermined. Military defeats in the previous month also
increase the likelihood of the selected democratic states changing their strategies. This
intuitive finding is reflected in the case studies as the United States is compelled to rescue
South Vietnam by a massive bombing campaign of the DRV. Likewise, the North
Korean victories over the ROK set the stage for an American maneuver strategy to entrap

the North Korean forces through the Inchon landing.

248

Political pressure is a comparatively robust and substantially significant finding.
Democratic political pressure variable also enjoys statistical significance at conventional
levels and a substantial impact on the likelihood of an attempt to change strategy. The
United States employment of a punishment strategy to drive Hanoi to the bargaining table
was in part reflective of the executive’s fear of the public and congress refirsing to
mobilize further. As political pressure jeopardized Nixon’s preferred war policies he
elected to employ to a brutal air campaign against the DRV with its attendant diplomatic
consequences.

The political pressure theory again offers a unified theory of why war policies
change. Instead of a focus upon how major democracies alter strategy on the basis of key
variables or that semi-autocratic states will take risk acceptant strategies, a more
comprehensive theory across regime types is presented. The political pressure theory
mitigates the importance of war selection and post-hoe review by assuming an intra-war
review of costs and benefits by domestic political actors and demonstrating the
heightened likelihood of strategy change based upon this review.

Whereas a change in how a war is conducted might serve as a tacit
acknowledgement of incompetent leadership by the executive, accordingly strategies
appear to be characterized by inertia. A superior strategy designed to mitigate the threat
to an executive’s political position can mean the adoption of a more efficient strategy.
Effectively the application of political pressure and a change in strategy can appreciably
alter the ultimate outcome of a war.

The objective characteristics indicating the comparative success of a state in a war

are meaningful in the democratic cases and less so in the autocratic sample. Of the above

249

hypotheses political pressure appears to have the most influence upon the likelihood of an
attempt at changing strategy. Even when objective factors such as casualties and
battlefield losses are accounted for, the effect upon the likelihood of attempted strategy
change is not especially strong. Instead as with change in issue-goods position, the
application of political pressure for an increase in a war’s utility gives impetus for an
alteration in tactics. The less than robust performance of the political pressure variable
still contributes as much to explicate changes in attempted strategy as any of the other
hypotheses evaluated in the autocratic cases models.

Both of the quantitative chapters illustrate the importance of political pressure in
catalyzing changes in war policy. An executive can be compelled to move away from
their preferred strategy, an infra-war assessment can therefore flmdamentally alter the
outcome of a conflict. Unlike the selection of which wars to fight the political pressure
theory offers expectations as to how states will conduct wars regardless of regime type.
Moreover, it also affords an explanation of state behavior in wars that is more reflective

of the historical record.

250

Chapter Six: Summary

Findings

The principal of convergence literature suggests that credible information
obtained through fighting, and perhaps negotiation offers, informs as to the relative
resolve and capabilities of combatants. A unitary assessment of the costs and benefits
results in a state’s decision to shift negotiation position (Bennett 2003). Does this
parsimonious model benefit from the added complexity of introducing domestic
institutions and audiences?

Fundamentally there is a need to reconcile the findings posited by the domestic
political institutions literature in international relations with the principal of convergence.
If states are in large part circumscribed in terms of what wars they fight and how they
fight them by their enfranchised constituents, then these factors should also dictate how a
state responds to an updated understanding of the balance of resolve and capabilities
between combatants.

As the institutional literature suggests it is the anticipation of sanction that
informs as to what kind of wars a state will undertake. If credible information regarding
the likely outcome of a war is provided to the executive, could this information also serve
to readjust domestic observers’ appreciation for a war as well?

The leader of a state is supposed to adhere to the wishes of his principals. If a
leader’s constituents acquiesce to war mobilization, but these domestic observes are party
to updated expectations of a war’s outcome then they might also alter their opinions of a

war. If the agent and principals assess updated information regarding a war differently,

251

then why would they suspend dissent with the executive’s choices until the very end of a
war?

The chain of reasoning that I use is that as a war progresses it is not just the
executive who updates expectations as to a war’s outcome. Instead autocratic elites and
democratic public have access to at least some of the same information that the leader of
a state does. The interests of a portion of these actors eventually become at odds with
current war policies relative to alternatives. Having the ability to deselect the executive
the interests of this group should be important to the leader of a state.

These updates occur as events dictate the relative balance and capabilities of the
opponents. The Tet Offensive and Midway were both deeply disconcerting to the
domestic political actors of the United States and Japan. Casualties helped drive down
the level of support for the Korean War and the prospect of revolution stirred up demands
for demobilization amongst the elites in the Islamic Republic. These costs were
juxtaposed to benefits less favorably by the principals of these regimes when compared to
the executive.

The political pressure theory offers expectations of how states will conduct wars
and the kind of negotiation offers they will pursue. As political pressure grows less
favorable issue-goods benefits are pursued and less costly strategies are employed. This
contrasts the more stark differentiation of autocracies being relatively free to pursue their
wars as they see fit and democratic executive having to contend with the prospect of
electoral sanction.

Autocracies can also have accountability over the executive. In as much as

differences of opinion are relatively concentrated when compared to a democracy,

252

demands for policy change emerge as threats to the interests of autocratic elites interests
emerge. The prospect of a regime’s collapse appears to be one of the principal drivers of
intense political pressure for both semi-autocracies and more totalitarian dictatorships.
However, even when an existential threat to a regime is not imminent political pressure
and changes in war policy can occur.

Three major means of altering war policies are available repression for autocratic
states, issue-goods position with the attendant level of mobilization, and alteration in
military strategy. Several hypotheses evaluating what changes a state’s appreciation for
what issue-goods they are likely to receive were tested. The variables indicating the level
of political endangerment for an executive appeared to enjoy statistical significance at
conventional levels across specifications. Substantially the likelihood of a lower issue-
goods negotiation position was evaluated against a number of alternative hypotheses.
However battles, capabilities, casualties, and other metrics of costs and benefits were
generally less important than the political pressure variables.

Empirical testing of the principal of convergence model relies heavily upon
casualties and total population to determine how states update their expectations
(Slantchev 2003a). A measure with greater validity takes a dataset of major battles and
their duration in days to address how states change their expectations of a war’s outcome
(Ramsay 2006). The battles dataset I employ is much more comprehensive than just
large and emblematic engagements.

My statistical analysis addresses the theoretical deficit of this literature by
showing how the impetus behind changes in issue-goods position is driven in part to

accommodate dissent with war policies. Reliance on just battlefield events or the costs

253

of fighting to determine how a war changes a state’s assessment of the likely outcome of
a war is incomplete, in as much as domestic political actors make assessments of an
ongoing war’s utility.

I also present a more comprehensive evaluation of why military strategies change.
Instead of focusing upon a set of key indicators or case-by-case explanations of what
changes battlefield tactics a more systematic approach is undertaken, a cross national
sample using consistent variables is employed (Gartner 1997). While operating a some
what higher level of abstraction than Gartner’s analysis, I confirm his major findings.
However, the substantial significance of the ratio of losses between a democracy and its
opponent are considerably less important than is the measure of political pressure.

The acquiescence for mobilization by domestic political actors is not the only
assessment of a war’s utility until a war concludes. An assessment of war policies in
terms of costs and benefits is made and if plausible alternatives to current war policies are
deemed superior demands for alternative policies follow. The case for the intra-war
assessment by domestic political actors is confirmed by both the quantitative and
qualitative analysis.

Imperial Japan’s decision to pursue a war was based upon a rationale calculation
of utility by the executive and ratified by the elites (Russett 1972). As it became clear
that American industrial might was matched by a stronger than anticipated resolve a
revision in previous expectations amongst Japan’s elite took place. Defeats at Midway
and Saipan altered domestic political actors’ expected utility of continuing to fight.

Battlefield events led to the deselection of two prime minister who would not

254

accommodate demands for changes in issue-goods position. Such that the endorsement
of war mobilization at the outset was replaced by dissent over war policies.

Changes in military strategy also mitigate the effects of political pressure. After
the Iraqi invasion of Iran was stopped and Khomeini began his war of attrition
apprehensions amongst Baghdad’s elites grew. As the costs of a potential revolution in
southern Iraq seemed plausible as Tehran’s forces pushed closer, Hussein altered
strategy. Establishing a defensive position to minimize casualties and using punishment
attacks the executive attempted to improve the utility calculation of elites.

To address political pressure repression can be employed. In the case of post-
revolutionary Iranian represents a case where the war refashions the size and composition
of elites. Early evidence of domestic political pressure in Iran emerged from the military
and leftist political elites that sought to contend for power. Even as purges took place
Khomeini held the course of an attritional strategy and demands for regime change in
Iraq. Even as the number of domestic political actors is winnowed down, political
pressure was still a major factor in the alteration of war policies in even the most

totalitarian systems.

Theoretical Contributions

The principal of convergence offers a theoretically robust and intuitively accurate
representation of how wars terminate. One major weaknesses of this research project is
that it offers few empirical insights into state behavior other than fighting provides for the

eventual termination of a conflict. For the most part this literature has offered a formal

255

construction of war onset and termination, but what informs states as to the relative
resolve and capabilities of the combatants remains undefined empirically.

As noted above, the empirical tests of the principal of convergence has been
limited and likely suffers from problems with the measurement validity. Because issue-
goods in most wars are multi-dimensional, different regimes weight casualties differently,
and the salience of a war varies massively across cases a measure that is endogenous to
the war in question is needed. An understanding of the comparative resolve and
capabilities of the opponents is best contextualized by observing battles.

The analysis of the principal of converge done above indicates that military
victories and defeats, have merit in determining moving the ongoing bargaining process
forward. But the empirical weight of political pressure, even when battles and casualties
are specified that greatly increases the likelihood of a less favorable negotiation offer and
consequently a war of reduced length. Theoretically this necessitates a restatement of the
assumptions of the model in order to accommodate the importance of domestic observers
of an ongoing war.

The conditions under which strategies changes occur has also been limited in its
scope. The literature has substantiated the importance of military strategy on war
outcome, but these studies treat strategy as constant over the duration of the war (Stam
1996; Reiter and Meek 1999). Other studies are qualitative focusing upon great power
democracies (Gartner 1997). The analysis of strategy change and political pressure as
identified the correlates of changes in strategy as well as the relative importance of these

variables.

256

The other literature that is addressed in my dissertation is the influence of
domestic politics conditions the selection and conduct of wars. Sanction is thought to
keep the executive attenuated to the wants of his/her constituents. The assessment of a
leader’s performance is assumed to be after a war has concluded. Similar to this intuitive
piece, Goemans’s theory implies that punishment should condition semi-autocratic states
to hazard riskier strategies in order to mitigate the threat of the executive’s demise. In
both cases it is a post-hoe review or within one year before the end of the war is assumed
to provide the verdict on an executive’s war policies.

In either case this presumes that the constituents of a state are unable to observe
the course of a war before hand. If it is only the public that is sheltered from this
information, than why would autocratic elites forestall their assessments? If internal
revolt is the concern, why would the populace of a state wait until the war is over and
costs are reduced in order to revolt?

Referencing another line of research, casualties are held to determine to a large
extent the public’s support for a war. The implication is that this assessment is made
while a state is still at war. The greater in duration and number of lives lost in a war, the
willingness to mobilize is reduced. While the costs of a conflict are in substantial part
attributable to war selection, an assessment by the public effectively alters the outcome of
a war. Instead of war termination being the only response to public discontent, an
executive could instead attempt to alter policies and perceptions of the public.

Likewise, the elites in autocracy likely face similar prospects for punishment
should the regime fall to foreign forces or to internal revolt. Anticipating the same

outcome that the executive of Goemans’s theory appraises, risk acceptant behavior might

257

be the only way to avoid the likelihood of punishment if the war progresses on its current
course. At which point the executive is now faced with the prospect of sanction unless
the inhibitions of his constituents are mitigated. Both democratic and autocratic
executives are theorized to respond with war policy changes in order to ensure their
political survival.

The implication is that the assumptions of a post-hoc review and a unitary
assessment of battlefield events are likely offering predictions of executive policies and
war outcomes that are less accurate than alternative models. Referencing the maj or wars
used in my analysis, it does not appear that aggressive strategies and heavier mobilization
are an effect of an executive’s anticipation of punishment, but instead it is the executive
that is compelled to undertake more conservative war policies. The importance of war
selection is also muted as the outcome of a war is anticipated well before hand, meaning
that war policies more in conformity to the wishes of the executive’s constituents are
employed.

Similarly, the democratic public’s casualty aversion can be mitigated by
improving the relative benefits of fighting. This can be accomplished by a number of
means including alteration of war policies and by exhibiting evidence of military progress
in statements encouraging further mobilization. By maintaining support for mobilization,
even if at the expense of reduced issue-goods position a democracy might not experience
the steady reduction in war fighting ability. In fact of the democratic cases sampled, no
meaningful correlation between war duration and the win percentage of states is found.

The political pressure theory demands more variables than regime type in order to

explicate a state’s war policies. However, once regime type is understood and the cost-

258

benefit schedule of the enfranchised constituency identified expectations as to strategy,
negotiation position as well as mobilization can be made.

In as much as executive policies are static and become inappropriate to the
changing conditions in a war, it is possible that political pressure might compel an
improvement in strategy. This is especially true of an autocracy where the propensity for
micromanagement as well as the profligate expenditure of lives and resources is more
common.

Democracies are likely more nuanced, in that casualties are found to be heavily
weighted in terms of the level of support for a war. Consequently the executive might be
forced into negotiation positions, strategies, and/or levels of mobilization that undercut
the ability of the state to maximize its security benefits. However, the restraint demanded
by the democratic public might reduce the likelihood of over extending a great power’s

security commitments (Snyder 1992).

Insights for Public Policy

The above results tacitly address a debate as to whether the influence of the public
in military affairs is relatively positive (Reiter and Stam 2002) or is more harmful to a
state’s interests (Kissinger 1994). The above results cannot be deemed conclusive,
however that relatively positive outcomes appear to come about due to political pressure
or deselection in a democratic context.

Upon the Republicans losing control of Congress in the 2006 mid-term elections
the executive took steps to alleviate domestic political pressure regarding war policies.

Secretary of Defense Donald Rumsfeld was considered a major impediment to change in

259

Iraq War policy and was replaced by Robert Gates. A new counter-insurgency strategy
was also employed under General David Petraeus, which substantially improved the
prospects for success in Iraq. Evidence of severe political pressure and a waning public
support for mobilization was addressed through a change in strategy.

Similarly, the relatively incremental and inflexible policy of Johnson in Vietnam
led to intense dissent with the war and eventually compelled the executive not to seek
reelection. The new executive demobilized the amount of resources being employed in
the war, reducing political pressure. Nixon also employed a more effective counter-
insurgency strategy under General Creighton Abrams. Nixon’s policies of diplomacy,
interdicting NVA supplies into the South, and eventually two punishing bombing
campaign garnered a modest peace settlement. Had his political fortunes not been
compromised due to Watergate he might have sustained the RVN even longer.

Conversely the American superpower is often assessed as being casualty sensitive
to the point of appearing unresolved, a sentiment which Osarna Bin Ladin has alluded to
in past statements (Record 2002). If perceptions of casualty aversion in democracies
pervades, this might invite future attacks in the belief that meaningful military action is
not forthcoming. However, the political pressure theory can also represent demands for
demobilization that can translate into more innovative war policies and restraint in the
scale of an ongoing commitment.

Autocracies appear to benefit from political pressure being exerted on the
executive. Micromanagement by Stalin and Hitler often led to disastrous results for their
armed forces. Once decisions were made by the professional military, albeit politically

reliable ones, generally the war was conducted with greater efficiency.

260

Evidence of domestic dissent is demonstrated to strengthen the hand of an
executive negotiating with its foreign opponent (Putnam 1988). This is found to be the
case because an executive must receive ratification of an agreement domestically. While
this is true of economic negotiations, likely the relatively transparent evidence of dissent
in a democracy and the more opaque political pressure of an autocracy undermine the

ability of a state to drive a hard bargain in peace negotiations.

Future Research Projects

With the data collected for my dissertation I have plans for several other projects.
The use of battles data from my research has been informative in helping to define what
offset the costs of casualties to the public in a democratic context. Instead of relying
upon experimental research and hypothetical questions, I plan to use battles as credible
indications of a war’s benefits (Feaver and Gelpi 2004; Boettcher and Cobb 2006).
Contextualizing benefits with battles permits an evaluation of the utility of the security
goods at stake in a war as well as the likelihood of achieving those aims.

Similarly a new literature can also be created that addresses questions of what
defines the autocratic domestic political actors' equivalence to the democracy casualty
literature. A more nuanced understanding of how elites in an autocracy assess a war’s
progress, which could contribute to an understanding of autocratic support for
mobilization.

Empirical evidence has been found that democracies are superior war fighters
(Reiter and Starn 2002). This finding has been challenged on the basis of being reliant

upon influential observations and on the comparative paucity of micro-foundational

261

causal factors (Desch 2003). I plan to examine the narrative of democratic public
wearying of the costs of a war and retracting support for mobilization, against alternative
hypotheses. With the use of my battles dataset and public opinion polling an analysis to
better define this relationship with more valid and detailed measures.

Another study would examine the prospect that democracies with their
comparatively transparent politics and well-known casualty aversion makes then targets
for manipulation by their opponents. Does the executive of an enemy state seek to
devalue the utility of the war to a comparatively powerful democracy through credible
signals? Using changes in strategy or signals of resolve, do the opponent’s of a
democracy have the opportunity and willingness to target public support?

In terms of a war it is found that casualties either cumulative (Mueller 1973) or
more recent battle deaths (Gartner and Segura 1998) help to define the level of support a
war enjoys. The salience of the war and evidence of progress have also been shown to
define benefits offsetting these costs (Gelpi and Feaver 2004). The number of opponent
casualties and the appearance of success all can influence public perceptions of a war
(Boettcher and Cobb 2006). The combination of costs and benefits are shown to craft

public opinion of a war.

262

References

Abuza, Zachary. 2001. Renovating Politics in Contemporary Vietnam: Politics in
Contemporary Vietnam. Boulder, CO: Lynne Rienner Publishers.

Allison, Graham. 1971. Essence of Decision: Explaining the Cuban Missile Crisis, 1"
edition Boston, MA: Little, Brown and Company.

Ambrose, Stephen E. 1989. Nixon the Triumph of a Politician 1962-1972. New York:
Simon and Schuster.

Ambrose, Stephen. 1995. D-Day June 6, 1944: The Climactic Battle of World War II.
New York: Simon and Schuster.

Ang, Cheng Guan. 2002. The Vietnam War fiom the Other Side: The Vietnamese
Communists' Perspective. London: Routledge

Ang, Cheng Guan. 2004. Ending the Vietnam War: The Vietnamese Communists'
Perspective. London: Routledge

Alperovitz, Gar. 1965. Atomic Diplomacy: Hiroshima and Potsdam. New York: Simon
and Schuster.

Avant, Deborah D. 1993. “The Institutional Sources of Military Doctrine: Hegemons in
Peripheral Wars.”, International Studies Quarterly, Vol. 37 (4): 409-430.

Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books.

Bateson, Charles. 1968. The War with Japan: A Concise History. East Lansing, MI:
Michigan State University Press.

Beaumont, Roger. 1987. “The Bomber Offensive as a Second Front” Journal of
Contemporary History, Vol. 22 (1): 3-19.

Beer, A. Francis and Thomas F. Mayer. 1986. “Why Wars End: Some Hypotheses”,
Review of International Studies, 12 (2): 95—106.

Bennett, D. Scott and Allan C. Starn. 1998. “The Declining Advantage of Democracy: A
Combined Model of War Outcomes and Duration”, Journal of Conflict Resolution, Vol.
42 (3): 344-366.

Bergarnini, David. 1971. Japan ’s Imperial Conspiracy. New York, NY: William
Morrow.

263

Bernstein, Barton J. 1975. “Roosevelt, Truman, and the Atomic Bomb, 1941-1945: A
Reinterpretation” Political Science Quarterly, Vol. 90(1): 23-69

Biddle, Stephen. 2004. Military Power: Explaining Victory and Defeat in Modern
Battle. Princeton, NJ: Princeton University Press.

Blainey, Geoffrey. 1973. The Causes of War. New York, NY: Free Press.
Blood, Jake. 2005. The Tet Effect: Intelligence and the Public Perception of War.
London: Routledge.

Boettcher, William A. and Michael D. Cobb. 2006. “Echoes of Vietnam: Casualty
F raming and Public Perceptions of Success and Failure in Iraq”, Journal of Conflict
Resolution 50(6): 831-854.

Botjer, George F. 1996. Sideshow War: The Italian Campaign, 1 943—1945. 1996,
College Station, TX: Texas A&M University Press.

Brackman, Roman. 2001. The Secret File of Joseph Stalin: A Hidden Life. London:
Frank Cass Publishers.

Broekmeyer, 2004. Stalin, the Russians, and Their War. Madison, WI: University of
Wisconsin Press.

Brooks, Lester. 1968. Behind Japan ’s Surrender: The Secret Struggle That Ended an
Empire. New York, NY: McGraw-Hill.

Bueno de Mesquita, Bruce and Randolph M. Siversion. 1995. “War and the Survival of
Political Leaders: A Comparative Study of Regime Types and Political Accountability”,
American Political Science Review, Vol. 89 (4): 841-855.

Bueno de Mesquita, Bruce, Alastair Smith, Randolph M. Siverson, and James D.
Morrow. 2003. The Logic of Political Survival. Cambridge, MA: MIT Press.

Bueno De Mesquita, Bruce. 1980. “An Expected Utility Theory of International
Conflict”, The American Political Science Review, Vol. 74 (4): 917-931.

Burns, James MacGregor. 1970. Roosevelt: The Soldier of Freedom. New York, NY:
Harcourt Brace J ovanovich.

Butow, Robert J .C. 1954. Japan’s Decision to Surrender. Stanford, CA: Stanford
University Press.

Butow, Robert J .C. 1961. T ojo and the Coming of the War. Princeton, NJ: Princeton
University Press.

264

Buzzanco, Robert. 1996. “The Myth of Tet: American Failure and the Politics of War”,:
231-258. in Marc Jason Gilbert and William Head eds. The Tet Offensive. Westport,
CT: Praeger.

Canes-Wrone, Brandice and Scott de Marchi. 2002. “Presidential Approval and
Legislative Success”, The Journal of Politics, 64 (2): 491—509.

Cantril, Hadley. 1948. The Public Opinion Quarterly, Vol. 12 (1): 30-44.

Casey, Stephen. 2008. The Korean War. New York: Oxford University Press.
Caridi, Ronald James. 1968. “The GOP. and the Korean War”, The Pacific Historical
Review, Vol. 37(4): 423-443.

Carland, John M. 2004. “Winning the Vietnam War: Westmoreland's Approach in Two
Documents”, The Journal of Military History, Vol. 68 (2): 553-574.

Chase, John L. 1955. “Unconditional Surrender Reconsidered” Political Science
Quarterly, Vol. 70(2): 258-279.

Chen, J ian. 1994. China 's Road to the Korean War: The Making of the Sim-American
Confiontation. New York: Columbia University Press.

Chen, J ian. 2001. Mao's China and the Cold War. Chapel Hill, NC: University of North
Carolina Press.

Chen, Xialu. 1989. “China’s Policy Toward the United States, 1949-1955.” In Harding,
Harry and Yuan Ming. eds. Sino-American Relations 1945-1955: A Joint Reassessment
of a Critical Decade. Wilmington, DE: SR Books.

Chen, Theodore H. E. 1960. Thought Reform of the Chinese Intellectuals. Hong Kong:
University of Hong Kong Press.

Chiozza, Giacomo and HE. Goemans. 2004. “International Conflict and the Tenure of
Leaders: Is War Still "Ex Post" Inefficient?”, American Journal of Political Science, Vol.
48 (3): 604-619.

Chubin, S. 1989. “Iran and the War: From Stalemate to Ceasefire”, in H. W. Maull and
0. Pick. eds. The Gulf War: Regional and International Dimensions. London: Pinter.

Chubin, S. and C. Tripp. 1988. Iran and Iraq at War. I.B. Tauris: London.

Clodfelter, Michael. 1995. Vietnam in Military Statistics: A History of the Indochina
Wars, 1 772-1991. Jefferson, NC: McFarland and Company.

Clodfelter, Michael. 1992. Warfare and Armed Conflicts: A Statistical Reference to
Casualty and Other Figures, 1618-1991. Jefferson, NC: Mcfarland & Co.

265

Clodfelter, Mark. 2006. The Limits of Air Power: The American Bombing of North
Vietnam. Lincoln, NE: University of Nebraska Press.

Coates, Tim. Editor. 2000. D Day to VE Day, 1944-45: General Eisenhower ’s Report
on the Invasion of Europe. London: The Stationery Office.

Colaresi, Michael. 2007. “The Benefit of the Doubt: Testing an Informational Theory of
the Rally-Effect”, International Organization, 61(1): 99-144.

Copeland, Dale C. 2000. The Origins of Major War. “Ithaca, NY: Cornell University
Press.

Cordesman, Anthony H. and AR. Wagner. 1991. The Lessons of Modern War, Volume
II: The Iran—Iraq War. Boulder, CO: Westview Press.

Cordesman, Anthony H. and Ahmed S. Hashim. 1997. Iraq: Sanctions and Beyond.
Boulder, CO: Westview Press.

Cotton, Timothy Y. C. 1986. “War and American Democracy: Electoral Costs of the
Last Five Wars”, The Journal of Conflict Resolution, 30 (4): 616-635.

Currey, Cecil B. 1996. Victory at Any Cost: the Genius of Viet Nam 's Gen. Vo Nguyen
Giap. Soho Press: New York, NY.

Currey, Cecil. 1996. “The Tet Offensive and Its’ Aftermath”,: 73-88 in Marc Jason
Gilbert and William Head eds. The Tet Oflensive. Westport, CT: Praeger.

Delaforce, Patrick. 2004. The Battle of the Bulge: Hitler's Final Gamble. London:
Longrnan Publishing.

DeWeerd, Harvey A. 1949. “The German Officer Corps Versus Hitler”, Military
Aflairs, Vol. 13 (4): 199-208.

Diehl, Paul F. 1992. “What are They Fighting For? The Importance of Issues in
International Conflict Research”, Journal of Peace Research, Vol. 29 (3): 333-344.

Downs, George W. and David M. Rocke. 1994. “Conflict, Agency and Gambling for
Resurrection: The Principal Agent Problem Goes to War”, American Journal of Political
Science, Vol. 38 (2): 362-380.

Dreyer, June Teufel. 1985. “Civil-Military Relations in the People's Republic of China.”
Comparative Strategy, Vol. 5 (1): 27-49.

Duffy, James P. and Vincent L. Ricci. 1992. Target Hitler: The Plots to Kill Adolf
Hitler. Westport, CT: Greenwood Publishing Group.

266

Duiker, William J. 1986. The Communist Road To Power In Vietnam. 2nd Edition
Boulder, CO: Westview Press.

Duiker, William. 1995. Sacred War: Nationalism and Revolution in a Divided Vietnam.
New York, NY: McGraw-Hill.

Duiker, William. 1996. The Communist Road to Power in Vietnam. Boulder, CO:
Westview Press.

Duiker, William. 2000. Ho Chi Minh. New York: Hyperion.
Dunn, Walter S, Jr. 1997. Kursk: Hitler’s Gamble, 1943. Greenwood Publishing Group.

Edwards, Paul M. 2006. Selling the Korean War: Propaganda, Politics, and Public
Opinion in the United States, 1950-1 953. Greenwood Publishing Group.

Erickson, John. 1975a. The Road to Stalingrad: Stalin’s War with Germany Vol. I. New
Haven, CT: Yale Univ. Press.

Erickson, John. 1975b. The Road to Berlin: Stalin ’s War with Germany Vol. II. New
Haven, CT: Yale Univ. Press.

Fearon, James D. 1994. “Domestic Audiences and the Escalation of International
Disputes”, American Political Science Review, Vol. 88 (3): 577-592.

F earon, James D. 1995. “Rationalist Explanations for War”, International Organization,
Vol. 49 (3): 379-414.

Fearon, James D. 1998. “Domestic Politics, Foreign Policy, and Theories of
International Relations.” Annual Review of Political Science, Vol. 1: 289-313.

Fearon, James D. 1998b. “Bargaining Enforcement and International Cooperation.”
International Organization 52 (1): 269-306.

Feaver, Peter D. and Christopher Gelpi. 2004. Choosing Your Battles: American Civil-
Military Relations and the Use of F orce. Princeton, NJ: Princeton University Press.

Filson, Darren and Suzanne Werner. 2002. “A Bargaining Model of War and Peace:
Anticipating the Onset, Duration and Outcome of War.” American Journal of Political
Science, Vol. 46(4): 819-838.

Foot, Rosemary J. 1990. A Substitute for Victory: the Politics of Peacemaking at the
Korean Armistice Talks. Ithaca, NY: Cornell University Press.

Foot, Rosemary J. 1988-1989. “Nuclear Coercion and the Ending of the Korean
Conflict”, International Security, Vol. 13 (3): 92-112.

267

Frank, Richard. 1999. Downfall: The End of the Imperial Japanese Empire.
New York, NY: Random House.

Fukai, Shigeko N. and Haruhiro Fukui. 1992. “Elite Recruitment and Political
Leadership”, Political Science and Politics, Vol. 25 (1): 25-36.

Gaddis, John Lewis. 1986. “The Long Peace: Elements of Stability in the Postwar
International System”, International Security, Vol. 10(4): 99-142.

Gallup and Fortune Polls. 1942. The Public Opinion Quarterly, Vol. 6 (1): 140-174

Gartner, Scott Sigmund and Gary M. Segura. 1998. “War, Casualties, and Public
Opinion”, Journal of Conflict Resolution, Vol. 42 (3): 278-300.

Gartner, Scott Sigmund and Marissa Edson Myers. 1995. “Body Counts and ‘Success’
in the Vietnam and Korean Wars”, Journal of Interdisciplinary History, Vol. 25 (3):
377-395.

Gartner, Scott Sigmund, Gary M. Segura, and Bethany A. Barratt. 2004. “War
Casualties, Policy Positions, and the Fate of Legislators”, Political Research Quarterly,
Vol. 57 (3): 467-477.

Gartner, Scott Sigmund. 1997. Strategic Assessment in War. New Haven, CT: Yale
University Press.

Gaubatz, Kurt Taylor. 1999. Elections and War: The Electoral Incentive in the
Democratic Politics of War and Peace. Stanford, CA: Stanford University Press.

Gelpi, Christopher, Peter D. Feaver and Jason Reifler. 2005/2006. “Success Matters:
Causality Sensitivity and the War in Iraq”, International Security, Vol. 30(3): 7-46.

Glantz, David M. 2005. Colossus Reborn: The Red Army at War, [941—1943.
Lawrence, KS: University Press of Kansas.

Gochman, Charles and Russell Leng. 1983. “Realpolitik and the Road to War: An
Analysis of the Attributes of Behavior”, International Studies Quarterly, Vol. 27 (l): 97-
120.

Goda, Norman J. W. 2000. "Black Marks: Hitler's Bribery of His Senior Officers
during World War II”, The Journal of Modern History, Vol. 72 (2): 413-452.

Goemans, HE. 2000. “Fighting for Survival: the Fate of Leaders and the Duration of
War”, Journal of Conflict Resolution, Vol. 44 (5): 555-579.

Goodman, Allan E. 1978. The Lost Peace: America ’s Search for a Negotiated Settlement
of the Vietnam War. Stanford, CA: Hoover Institution Press.

268

Gurtov, Melvin and Byong-Moo Hwang. 1980 China under Threat: The Politics of
Strategy and Diplomacy. Baltimore, MD: Johns Hopkins University Press.

Halliday, Fred. 1980. “Iran's Revolution: The First Year”, MERIP Reports, No. 88: 3-5.

Halperin, Morton H. 1963. “The Limiting Process in the Korean War”, Political Science
Quarterly, Vol. 78 (1): 13-39.

Hamby. Alonzo L. 1992. Liberalism and Its Challengers: From F.D. R. to Bush. 2“d Ed.
New York, NY: Oxford University Press.

Harnrin, Carol Lee. 1994. “Elite politics and the Development of China's Foreign
Relations”: 70-114. In Thomas W. Robinson and David Shambaugh. eds. Foreign
Policy: Theory and Practice. New York, NY: Oxford University Press.

Hastings, Max. 2007. Retribution: The Battle for Japan, 1944-45. New York: Harper
Press.

Healy, Mark. 1992. Kursk 1943: Tide T urns in the East. Osprey Publishing.

Herzstein, Robert E. 1979. The War That Hitler Won: The Most Infamous Propaganda
Campaign in History. London: Hamish Hamilton.

Herzog, Chaim. 1989. “A Military-Strategic Overview.” in The Iran-Iraq War: Impact
and Implications. Efraim Karsh ed. London: Macmillan.

Higgins, Trumbull. 1970. “The Anglo-American Historians' War in the Mediterranean,
1942-1945” Military Aflairs, Vol. 34 (3): 84-88.

Hiro, Dilio. 1984. “The Strange War in the Gulf”, MERIP Reports, No. 125/ 126: 3-14.

Hiro, Dilip. 1985. Iran Under the Ayatollahs. London: Routledge & Kegan Paul PLC.

Hiro, Dilip. 1987. Iran Under the Ayatollahs. 2““ ed. London: Routledge & Kegan Paul
PLC.

Hiro, Hiro. 1991. The Longest War: The Iran-Iraq Military Conflict. London: Routledge
& Kegan Paul PLC.

Hoffman, Peter. 1988. German Resistance to Hitler. Cambridge, MA: Harvard
University Press.

Holsti, Kalevi. 1991. Peace and War: Armed Conflicts and International Order I 648-
1989. New York: Cambridge University Press.

269

Hooglund, Eric. 1989. “Iran’s Revolution Turns Ten.” Middle East Report, Vol. 156: 4-
12.

Hoyt, Edwin P. 1988. Hitler's War. New York: McGraw-Hill.

Huang, Jing. 2000. Factionalism in Chinese Communist Politics. New York:
Cambridge University Press.

Huth, Paul K. and Todd L. Allee. 2002. The Democratic Peace and Territorial Conflict
in the Twentieth Century. New York: Cambridge University Press.

Hyde, H. Montgomery. 1971. Stalin: the History of a Dictator. New York: Farrar,
Straus and Giroux.

Ike, N. 1967. Japan 's Decision for War. Stanford, CA: Stanford University Press.
Irving, David. 1977. Hitler’s War. New York: Viking Press.

Jackson, W. G. F. 1967. The Battle for Italy. New York: Harper and Row.

Jacobs, Lawrence R. and Robert Y. Shapiro. 1999. "Lyndon Johnson, Vietnam, and
Public Opinion: Rethinking Realist Theory of Leadership." Presidential Studies
Quarterly, Vol. 29 (3): 592-616.

J aggers, Keith and Ted Robert Gurr. 1995. “Tracking Democracy's Third Wave with the
Polity III Data”, Journal of Peace Research, Vol. 32 (4): 469-482.

Jansen, Marius B. 2000. The Making of Modern Japan. Cambridge, MA: Harvard
University Press.

Jeffrey Kimball. 2004. The Vietnam War Files: Uncovering the Secret History of Nixon-
Era Strategy. Lawrence, KS: University Press of Kansas.

Jencks, HW. 1982. Muskets to Missiles: Politics and Professionalism in the Chinese
Army, 1945-1981. Boulder, CO: Westview Press.

Jervis, Robert. 1980. “The Impact of the Korean War on the Cold War”, Journal of
Conflict Resolution, Vol. 24 (4): 563-592.

Karnow, Stanley. 1972. Mao and China: From Revolution to Revolution. New York:
Viking Press. ‘

Karsh, E. and I. Rautsi 1991. Saddam Hussein: A Political Biography. London: Kings
College.

270

Kase, Toshikazu. 1950. Journey to the Missouri. New Haven, CT: Yale University
Press.

Kennedy, Paul M. 1987. The Rise and Fall of the Great Powers: Economic Change and
Military Conflict. New York: Random House.

Khong, Yuen Foong. 1993. Analogies at War: Korea, Munich, Dien Bien Phu, and the
Vietnam Decision of 1965. Princeton, NJ: Princeton University Press.

Kimball, Jeffiey P. 1998. Nixon's Vietnam War. Lawrence, KS: University Press of
Kansas.

Kimball, Warren F. 1996. “Stalingrad: A Chance for Choices” The Journal of Military
History, Vol. 60 (1): 89-114.

King, Gary and Langche Zeng. 2001. "Logistic Regression in Rare Events Data,"
Political Analysis, Vol. 9 (2): 137-163.

King, Gary and Langche Zeng. 2002. “Estimating Risk and Rate Levels, Ratios, and
Differences in Case-Control Studies.” Statistics in Medicine, Vol. 21: 1409-1427.

Tomz, Michael, Gary King, and Langche Zheng. 1999. RELOGIT: Rare Events Logistic
Regression, Version 1.1 . Cambridge, MA.

Kissinger, Henry. 1994. Diplomacy. New York, NY: Simon and Schuster.

Klein, Donald W. and Anne B. Clark. 1971. Biographic Dictionary of Chinese
Communism, 1921-1965. Vol. 11. Cambridge, MA: Harvard University Press.

Knox, M. 2000. “Wehrmacht Officer Policy and Social Revolution”, The Historical
Journal, Vol. 43: 801-825.

Koen, Ross Y. 1974. The China Lobby in American Politics. New York:

Korean Institute of Military History. 2001a. The Korean War, Vol. 1. Lincoln, NE:
University of Nebraska Press.

Korean Institute of Military History. 2001b. The Korean War, Vol. 11. Lincoln, NE:
University of Nebraska Press.

Korean Institute of Military History. 200C]. The Korean War, Vol. 111. Lincoln, NE:
University of Nebraska Press.

Larson, E. V. 1996. Casualties and consensus: The historical role of casualties in
domestic support for US. military operations. Santa Monica, CA: RAND.

271

Leslie H. Gelb, Richard K. Betts. 1979. The Irony of Vietnam: The System Worked.
Washington, DC: Brookings Institution Press.

Levi, Ariel and Philip E. Tetlock. 1980. “A Cognitive Analysis of Japan's 1941
Decision for War”, Journal of Conflict Resolution, Vol. 24 (2): 195-211.

Li, Xiaobing, Allan R. Millett and Bin Yu. 2001. Mao ’s Generals Remember Korea.
Lawrence, KS: University of Kansas Press.

Levy, David W. 1995. The Debate over Vietnam. 2"“ edition Baltimore, MD: Johns
Hopkins University Press.

Levy, Jack. 1989. “The Diversionary Theory of War” in Mandus Midlarski ed.
Handbook of War Studies. London: Unwin Hyman: 259-288.

Lunch, William L. and Peter W. Sperlich. 1979. “American Public Opinion and the War
in Vietnam”, The Western Political Quarterly, Vol. 32 (1): 21-44.

MacFarquhar, Roderick. 1997. The Politics of China: The Eras of Mao and Deng.
New York: Cambridge University Press.

Mackey, Sandra. 1996. The Iranians: Persia, Islam and the Soul of a Nation.
New York: Dutton.

Magenheimer, Heinz. 1999. Hitler's War: Germany's Key Strategic Decisions 1940-
1945. London: Wellington House.

Malkasian, Carter. 2002. A History of Modern Wars of Attrition. Wesport, CT: Praeger.

Malkasian, Carter. 2004. “Toward a Better Understanding of Attrition: The Korean and
Vietnam Wars”, The Journal of Military History, Vol. 68 (3) p. 911-942.

Mao, Zedong. 1963. Selected Military Writings. Beijing: Foreign Languages Press.

Maoz, Zeev. 1989. “Power, Capabilities, and Paradoxical Conflict Outcomes.”
World Politics, Vol. 41 (2): 239-266

Martin, Bradley K. 2004. Under the Loving Care of the F atherly Leader: North Korea
and the Kim Dynasty. New York: Thomas Dunne.

Mastrry, Vojtech. 1972. “Stalin and the Prospects of a Separate Peace in World War II”,
The American Historical Review, Vol. 77 (5): 1365-1388.

272

 

McCullough, David 1992. Truman. New York, NY: Simon and Schuster.

McGinnis, Michael. 1986. “Issue Linkage and the Evolution of International
Cooperation”, Journal of conflict Resolution, Vol. 30 (1): 141-170.

Mearsheimer, John J. 2001. The Tragedy of Great Power Politics. New York, NY: -W.
W. Norton & Company.

Mearsheimer, John. 1999. “Hitler and Blitzkrieg Strategy”, in Robert Art and Kenneth
Waltz The Use of Force: Military Power and International Politics. 5“1 Edition New
York, NY: Rowan and Littlefield Publishers.

Mendelson, Sarah E. 1993. “Internal Battles and External Wars: Politics, Learning and
the Soviet Withdrawal from Afghanistan”, World Politics Vol. 45 (3): 327-360.

Milani, Mohsen M. 1994. The Making of Iran 's Islamic Revolution: From Monarchy to
Islamic Republic. Boulder, CO: Westview Press.

Mitcham, Samuel W. Jr. 1984. Rommel ’s Desert War: The Life and Death of the Afiika
Korps. New York, NY: Stein and Day.

Mitcham, Samuel W. Jr. 1984. Triumphant Fox: Erwin Rommel and the Rise of the
Afiika Korps. New York, NY: Stein and Day.

Moin, Bager. 1988. “Questions of Guardianship in Iran”, Third World Quarterly, Vol.
10(1): 191-200.

Mommsen, Hans. 1992. “The German Resistance against Hitler and the Restoration of
Politics.” The Journal of Modern History Vol. 64 p. 81 12-8127.

Morgan, T. Clifton and Sally Howard Campbell. 1991. “Domestic Structure, Decisional
Constraints and War: So Why Kant Democracies F ight?”, Journal of Conflict Resolution,
Vol. 35 (2): 187-211. .

Morgan, T. Clifion. 1994. Untying the Knot of War: A Bargaining Theory of
International Crises. (Univ. of Michigan Press: Ann Arbor, MI).

Mott, William H. IV and Jae Chang Kim. 2006. The Philosophy of Chinese Military
Culture .' Shih vs. Li. Palgrave Macmillan: New York, NY.

Mueller, John. 1973. War, Presidents and Public Opinion. (Wiley: New York)

Nasr, Vali. 2006. The Shia Revival: How Conflicts within Islam Will Shape the Future.
New York: W.W. Norton and Company.

Newton, Steven H. 2003. Kursk: The German View: Eyewitness Reports of Operation
Citadel by the German Commanders. Cambridge, MA: Da Capo Press, 2003.

273

 

Nikki R. Keddie and Eric Hooglund. 1986. Editors. The Iranian Revolution and the
Islamic Republic. Syracuse University Press: Syracuse NY.

Nobecourt, Jacques. 1967. Hitler 's Last Gamble: The Battles of the Ardennes.
Translated by R H Barry. London: Chatto & Windus.

Nordhaus, William D. 2002. “The Economic Consequences of a War with Iraq.”
Unpublished Manuscript.

Norpoth, Helmut. 1996. “Presidents and the Prospective Voter”, The Journal of
Politics, Vol. 58 (3): 776-792.

Norton, Edrvard C., Hua Wang and, Chunrong Ai. 2004. “Computing Interaction
Effects and Standard Errors in Logit and Probit Models”, The Stata Journal, Vol. 4 (2):
754-767.

O’Balance, Edgar. 1988. The Gulf War. New York, NY: Brassey’s Defence Publishers.

Oberdorfer, Don. 1971. Tet! New York, NY: Doubleday.

Olson, Mancur. 1982. The Rise and Decline of Nations: Economic Growth, Staglation,
and Social Rigidities. New Have, CT: Yale University Press.

Ostrom, Charles W. Jr. and Brian L. Job. 1986. “ The President and the Political Use of
Force”, The American Political Science Review, Vol. 80 (2): 541-566.

Pape, Robert. 2005. Dying to Win: The Strategic Logic of Suicide Terrorism. New York,
NY: Random House.

Parrish, Michael. 2004. Sacrifice of the Generals: Soviet Senior Oflicer Loses, I939-
I953. Lanham, MD: Scarecrow Press.

Peceny, Mark, Caroline C. Beer and Shannon Sanchez-Terry. 2002. “Dictatorial
Peace?”, American Political Science Review, Vol. 96 (1): 15-26.

Pickering, Jeffrey. 2002. “War-Weariness and Cumulative Effects: Victors,

Vanquished, and Subsequent Interstate Intervention”, Journal of Peace Research, Vol. 39
(3): 313-337.

Pierpaoli, Paul G. 1999. Truman and Korea The Political Culture of the Early Cold
War. Columbia, MO: University of Missouri Press

Pike, Douglas. 1989. “Origins of Leadership Change in the Socialist Republic of
Vietnam”: 107-128. in Ray Taras and Raymond Taras. Leadership Change in
Communist States. London: Routledge.

274

 

Pillar, Paul R. 1983. Negotiating Peace: War Termination as a Bargaining Process.
Princeton University Press: Princeton, NJ.

Posen, Barry R. 1984. The Sources of Military Doctrine: France, Britain and Germany
Between the world Wars. Ithaca, NY: Cornell University Press.

Posen, Barry R. and Andrew L. Ross. 1996. “Competing Visions for US Grand
Strategy”, International Security, Vol. 21 (3): 5-53.

Potter, EB. and Chester W. Nimitz Eds. 1960. The Great Sea War: The Story of Naval
Action in World War II. New York, NY: Bramhall House.

Powell, Ralph L. 1968. “Maoist Military Doctrines”, Asian Survey, Vol. 8 (4): 239-262.
Prados, John. 2004. “The Shape of the Table: Nguyen Van Thieu and Negotiations to
End the Conflict.”: 355-3 70 in Lloyd C. Gardner and Ted Gittinger eds. The Search for
Peace in Vietnam, 1964-1968. College Station, TX: Texas A&M University Press.
Preston, Andrew. 2004. “Missions Impossible: Canadian Secret Diplomacy and the
Quest for Peace in Vietnam”: 117-143 in Lloyd C. Gardner and Ted Gittinger eds. The

Search for Peace in Vietnam, 1964-I968. College Station, TX: Texas A&M University
Press.

Public Opinion Polls. 1944. The Public Opinion Quarterly, Vol. 8 (2): 276-303.
Quarter's Polls Source. 1945. The Public Opinion Quarterly, Vol. 9(1): 83-109.

Quinnett, Robert L. 1978. “The German Army Confronts the NSFO”, Journal of
Contemporary History, Vol. 13 (1): 53-64.

Ramsay, Kristopher W. 2006. “Settling it on the Field: Battlefield Events and War
Termination.” Presentation at the Midwest Political Science Meeting.

Randolph, SP. 2007. Powerful and Brutal Weapons: Nixon, Kissinger, and the Easter
Offensive. Cambridge, MA: Harvard University Press.

Rayfield, Donald. 2004. Stalin and His Hangmen. London: Viking/Penguin.

Razi, G. Hossein. 1988. “An Alternative Paradigm to State Rationality in Foreign
Policy: The Iran-Iraq War”, The Western Political Quarterly, Vol. 41 (4): 689-723.

Record, J. 2002. “Collapsed Countries, Casualty Dread, and the New American Way of
War.” Parameters. US Army War College.

275

Reissner, Johannes 1989. “The Iranian Revolution and the Iran-Iraq War”, in The Gulf
War: Regional and International Dimensions, edited by H. W. Maull and 0. Pick.
London: Pinter.

Reiter, Dan and Alan C Stam. 2002. Democracies at War. Princeton, NJ: Princeton
University Press.

Reiter, Dan and Curtis Meek. 1999. “Determinants of Military Strategy, 1903-1994: A
Quantitative empirical Test”, International Studies Quarterly Vol. 43 (2): 363-387.

Robert Powell. 2004. “Bargaining and Learning While Fighting ” American Journal of
Political Science, Vol. 48 (2): 344—361.

Roberts, Geoffrey. 2007. Stalin ’s Wars: From World War to Cold War, 1939-1953.
New Haven, CT: Yale University Press.

Rogers, Robert F. 1976. “Policy Differences in the Hanoi Leadership.” Studies in
Comparative Communism, spring/summer 1976

Rotundo, Louis. 1989. “Stalin and the Outbreak of War in 1941.” Journal of
Contemporary History, Vol. 24 (2): 277-299.

Rouleau, Eric. 1981. “The War and the Struggle for the State.” MERIP Reports, Vol. 98:
3-8.

Russett, Bruce and Alfred Stepan. Editors. 1973. Military Force and American Society.
(Harper & Row: New York).

Russett, Bruce M. “ Pearl Harbor: Deterrence Theory and Decision Theory.”

Russett, Bruce M. 1990. Controlling the Sword: The Democratic Governance of
National Security. Cambridge, MA: Harvard University Press.

Schreiber, EM. 1973. The Public Opinion Quarterly, Vol. 37 (1): 91-98.

Schreiber, E. M. 1976. “Anti-War Demonstrations and American Public Opinion on the
War in Vietnam”, The British Journal of Sociology, Vol. 27 (2): 225-236.

Schultz, Kenneth A. 1999. “Do Democratic Institutions Constrain or Inform?
Contrasting Two Institutional Perspectives on Democracy and War”, International
Organization, Vol. 53 (2): 233-266

Scobell, Andrew. 2003. China ’s Use of Military Force: Beyond the Great Wall and the
Long March. Cambridge University Press. Cambridge, UK

Segal, Gerald. 1985. Defending China. New York, NY: Oxford University Press.

276

Shillony, ben-Ami. 1981. Politics and Culture in Wartime Japan. Oxford: Clarendon
Press.

Shultz, Richard. 1978. “Breaking the Will of the Enemy during the Vietnam War: The
Operationalizatin of the Cost-Benefit Model of Counterinsurgency Warfare.” Journal of
Peace Research, Vol. 15 (2): 109-129

Skates, John Ray. 1994. The Invasion of Japan: Alternative to the Bomb. Columbia, SC:
University of South Carolina Press.

Slantchev, Branislav L. 2003a. “The Principle of Convergence in Wartime
Negotiations.” American Political Science Review Vol. 97 (4): 621-632.

Slantchev, Branislav L. 2003b. “The Power to Hurt: Costly Conflict with Completely
Informed States.” American Political Science Review Vol. 97 (1): 123-133.

Slantchev, Branislav L. 2004. “How Initiators End Their Wars: The Duration of

Warfare and the Terms of Peace.” American Journal of Political Science Vol. 48 (4):
813-829.

Small, Melvin and Singer, J. David. 1982. Resort to Arms: International and Civil Wars,
1816-1980. Beverly Hills, CA: Sage Publications.

Smith, Alastair. 1998. “Fighting Battles, Winning Wars”, The Journal of conflict
Resolution, Vol. 42 (3): 301-320.

Snyder, Glen H. and Paul Diesing. 1977. Conflict Among Nations. Princeton, NJ:
Princeton University Press.

Sorley, Lewis. 1999. A Better War: The Unexamined Victories and Final T rageay of
America ’s Last Years in Vietnam. New York, NY: Harcourt Brace.

Stam, Alan C. 1996. Win, Lose or Draw. Ann Arbor, MI: University of Michigan Press.
Steele, Richard W. 1978. “American Popular Opinion and the War Against Germany:

The Issue of Negotiated Peace, 1942” The Journal of American History, Vol. 65 (3): 704-
723

Stork, Joe. 1981. “Iraq and the War in the Gulf”, MERIP Reports, Vol. 97: 3-18

Strawson, John. 1969. The Battle for North Afiica. New York, NY: Charles Scribner’s
Sons.

Strawson, John. 1972. The Battle for the Ardennes. London: B.T.Batsford.

277

 

Strunk, Mildred. 1951. The Public Opinion Quarterly, Vol. 15 (2): 381-400.

Stueck, William. 2002. Rethinking the Korean War: A New Diplomatic and Strategic
History. Princeton, NJ: Princeton University Press.

Sullivan, Lawrence R. 1986—1987. “Leadership and Authority in the Chinese
Communist Party: Perspectives from the 19505.” Pacific Aflairs, Vol. 59 (4): 605-633.

Taylor, Telford. 1958. The March of Conquest: The German victories on Western
Europe, 1940. New York, NY: Simon and Schuster.

Teiwes, Frederick C. 1978. Elite Discipline in China: Coercive and Persuasive
Approaches to Rectification, 1 950-1953. Canabera: Austrailian National University
Press.

Teiwes, Frederick C. 1990. Politics at Mao ’s Court: Gao Gang and Party F actionalism
in the Early I950s. New York, NY: ME. Sharpe.

Teiwes, Frederick C. 1993. Politics and Purges in China. Rectification and the Decline
of Party Norms. 2"‘1 edition. Amok, NY. ME Sharpe.

Thai, Quang Trung. 1985. Collective Leadership and Factionalism. Singapore:
Institute of Southeast Asian Studies.

Thornton, Richard C. 2000. Odd Man Out: Truman, Stalin, Mao, and the Origins of the
Korean War. London: Brassey's.

Toland, John. 1970. The Rising Sun: The Decline and Fall of the Japanese Empire
1936-1945. New York, NY: Random House.

Tomz, Michael, Gary King, and Langche Zeng. 1999. RELOGIT: Rare Events Logistic
Regression, Version 1.1 Cambridge, MA: Harvard University.

Tripp, Charles. 1993. “The Iran-Iraq War and the Iraqi State.” in Iraq: Power and
Society. Derek Hopwood, Habib [show and Thomas Koszinowski eds. Oxford: Ithaca
Press.

Turley, William S. 1969. “Civil- Military Relations 1n North Vietnam”, Asian Survey,
Vol. 9 (12). 879-899.

Uhalley, Stephen. 1988. A History of the Chinese Communist Party. Stanford, CA:
Hoover Institution Press.

Van Dyke, Jon M. 1972. North Vietnam ’s Strategy for Survival. Palo Alto, CA: Pacific
Books.

278

 

Van Evera, Stephen. 1984. “The Cult of the Offensive and the Origins of the First
World War”, International Security, Vol. 9(1): 58-107.

Verba, Sidney and Richard A. Brody. 1970. “The Korean War”, The Public Opinion
Quarterly, Vol. 34(3): 325-332.

Voegelin, Eric. 1987. The New Science of Politics: An Introduction. Chicago, IL:
University of Chicago Press.

Wagner, R. H. 2000. “Bargaining and War”, American Journal of Political Science,
Vol. 44 (3): 469-484.

Walker, Stephen G. and Mark Schafer. “The Political Universe of Lyndon B. Johnson
and His Advisors: Diagnostic and Strategic Propensities in Their Operational Codes”,
Political Psychology, Vol. 21 (3): 529-543.

Walt, Stephen M. 1987. The Origins of Alliances. New York, NY: Cornell University
Press.

Walt, Stephen M. 1992. “Revolution and War,” World Politics, Vol. 44 (3):

Waltz, Kenneth N. 195 9. Man, the State and War. New York, NY: Columbia University
Press.

Watson, D. 2005. Molotov: a Biography. New York, NY: Palgrave, Macmillan,
Basingstoke.

Wells, Tom. 1994. The War Within: America’s Battle over Vietnam.
New York: Henry Holt and Company.

Wenger, Martha and Dick Anderson. 1987. “The Gulf War: Re-Flagging the Gulf”
MERIP Middle East Report, No. 148, (Sep-Oct): 23-26.

Weil, Herman M. 1975. “Can Bureaucracies Be Rational Actors? Foreign Policy

Decision-Making in North Vietnam”, International Studies Quarterly, Vol. 19 (4): 432-
468.

West, Phillip. 1992. “The Korean War and the Criteria of Significance in Chinese
Popular Culture.” The Journal of American-East Asian Relations Vol. 1 (4): 383-408.

Whiting, Allen S. 1960. China Crosses the Yalu: The Decision to Enter the Korean
War. New York, NY: MacMillan.

Wilmot, Chester. 1952. The Struggle for Europe. New York, NY: Harper & Brothers.

Wint, Guy. 1960. “China and Asia.” The China Quarterly, Vol. 1: 61-71.

279

Wittman, Donald. 1979. “How a War Ends: A Rational Model Approach”, The Journal
of Conflict Resolution, Vol. 23 (4): 743-763.

Wright, Claudia. 1985. “Religion and Strategy in the Iraq-Iran War” Third World
Quarterly, Vol. 7 (4): 839-852.

Wu, Chen-Tsai. 1970. Chinese Communist Who ’s Who. Taibei, Taiwan: Institute of
International Relations.

Xia, Yafeng. 2006. Negotiating with the Enemy: U.S.-China Talks During the Cold
War, 1949-1972. Bloomington, IN: Indiana University Press.

Yanaga, Chitoshi. 1949. Japan Since Perry. New York, NY: McGraw-Hill.

Yearbook on International Communist Afi'airs: 1966. 1967. Stanford, CA: Hoover
Institution Press. s.v. “North Vietnam.”

Yearbook on International Communist Affairs: 1968. 1969. Stanford, CA: Hoover
Institution Press. s.v. “North Vietnam.”

Yearbook on International Communist Aflairs: 1969. 1970. Stanford, CA: Hoover
Institution Press. s.v. “North Vietnam.”

Yearbook on International Communist Aflairs: 1 970. 1971. Stanford, CA: Hoover
Institution Press. s.v. “North Vietnam.”

Yearbook on International Communist Aflairs: 1972. 1973. Stanford, CA: Hoover
Institution Press. s.v. “North Vietnam.”

Yearbook on International Communist Aflairs: 1973. 1974. Stanford, CA: Hoover
Institution Press. s.v. “North Vietnam.”

Yelton, David K. 2000. "Ein Volk Steht Auf: The German Volkssturm and Nazi
Strategy, 1944-45”, The Journal of Military History, Vol. 64 (4): 1061-1083.

Young, Marilyn B. 1991. The Vietnam Wars: I945-1990. New York: Harper Perennial.
Zabih, S. 1988. The Iranian Military in Revolution and War. London: Routledge.

Zhang, Shu Guang. 1995. Mao 's Military Romanticism: China and the Korean War,
1950-1953. Lawrence, KS: University Press of Kansas.

Zhu, Pingchao. 2001. Americans and Chinese at the Korean War Cease-F ire
Negotiations, 1950-I953. Lewiston, NY: Edwin Mellen Press.

280

 

 

          

MCI MIL .AH SIA NM HSIIV LHIIIHAI ILyS

I III II III II I I II IIII III I I

1293 02956 9369

r_-