Contributions to Economics

Lucía Martínez Ordóñez

Military
Operational
Planning and
Strategic Moves
Contributions to Economics
More information about this series at http://www.springer.com/series/1262
Lucía Martínez OrdóQnez

Military Operational
Planning and Strategic
Moves

123
Lucía Martínez OrdóQnez
Faculty of Economics
Ruhr University Bochum
Bochum, Germany

ISSN 1431-1933 ISSN 2197-7178 (electronic)

Contributions to Economics
ISBN 978-3-319-56107-3 ISBN 978-3-319-56108-0 (eBook)
DOI 10.1007/978-3-319-56108-0

Library of Congress Control Number: 2017940402

© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, express or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To the memory of my dearest friend,
Ilka Reichel, who I know would have been
very happy to celebrate the finalization of
this work with me.
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 1
2 Game Theory and the Decision-Making Process
in Military Affairs.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 3
2.1 Game Theory: A Theory of Interdependent Decision .. . . . . . . . . . . . . . . . . 3
2.2 Some Comments on Using Game Theory to Model Tactics . . . . . . . . . . . 5
3 Strategies and Tactics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 11
3.1 A Brief History of Military Thought .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 11
3.2 Doctrine and the Organizational Levels of Conflicts . . . . . . . . . . . . . . . . . . . 16
3.3 Some War Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 17
3.4 Network-Centric Warfare .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 19
4 Conflicts as Zero-Sum Games . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 25
4.1 Asymmetric Conflicts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 25
4.2 Zero-Sum Games.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 29
4.3 Max-Min Strategies and Mixed Strategies Equilibrium
with Ordinal Payoffs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 32
5 The Advantage of Moving First Versus a First-Mover Advantage . . . . . . 37
5.1 Anti-Access/Area Denial (A2/AD).. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 38
5.2 Mining .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 40
5.3 A2/AD and the China Seas: A Wider Perspective .. . . . . . . . . . . . . . . . . . . . . 43
6 Irregular Warfare .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 49
6.1 Aspects of Irregular Warfare . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 50
6.2 Defining Irregular Warfare? Some Comments
on Objectives in Irregular Warfare . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 55
6.3 Designing Operations in Counterinsurgency Warfare .. . . . . . . . . . . . . . . . . 56

vii
viii Contents

7 Modelling Specific Aspects of COIN Operations:

On the Possibility of a First Mover Advantage . . . . . . .. . . . . . . . . . . . . . . . . . . . . 59
7.1 “The Surge” Versus a Surge . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 59
7.1.1 “The Surge” .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 63
7.1.2 The Model.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 64
7.2 “Clear Hold and Build” vs “Search and Destroy” .. . . . . . . . . . . . . . . . . . . . . 66
7.2.1 The Model.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 68
7.3 “Clear, Hold and Build” and “The Surge” . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 71
7.4 Surge: An Enemy-Centric Approach .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 72
7.4.1 Simultaneous Game . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 74
7.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 78
8 The War Against the Taliban: Tactical Operations
and Strategic Moves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 81
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 81
8.2 Scenarios and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 83
8.3 Solving the Simultaneous Game . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 85
8.4 Reality Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 87
8.5 A Wider View: Strategic Moves?.. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 88
8.6 Concluding Remarks .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 89
9 Lessons Learned . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 91

Acknowledgements.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 93

References .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 95
List of Abbreviations

A2/AD Anti-Access/Area Denial

AQI Al-Qaeda Iraq
BCA Budget Control Act
CAS Close Air Support
CHB Clear, Hold and Build
COIN Counterinsurgency
DoD Department of Defense
EDA European Defence Agency
EU European Union
FAC Forward Air Controller
FM 3-24 Field Manual 3-24: Counterinsurgency
FOB Forward Operating Bases
GWS Guerrilla Warfare Strategy
IDA Institute for Defense Analyses
IED Improvised Explosive Device
ISAF International Security Assistance Force
ISIS Islamic State of Iraq and Al-Sham
ISS Institute for Security Studies
IW Irregular Warfare
JAM-GC Joint Concept for Access and Maneuver in the Global Commons
JP 3-24 Joint Publication 3-24: Counterinsurgency Operations
JTAC Joint Terminal Air Controller
NATO North Atlantic Treaty Organization
NCW Network-Centric Warfare
OCO Overseas Contingency Operations
OFT Office of Force Transformation
OOTW Operations Other Than War
PSC Private Security Contractors
RRC Raising Rivals’ Costs
S&D Search and Destroy
SACEUR Supreme Allied Commander Europe

ix
x List of Abbreviations

UCDP Uppsala Conflict Data Program

WWI First World War
WWII Second World War
List of Figures

Fig. 4.1 Payoff matrix for the strategic interaction model .. . . . . . . . . . . . . . . . . . . . . 28

Fig. 4.2 Payoff matrix for zero-sum games . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 30
Fig. 4.3 Payoff matrix for case i) . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 31
Fig. 4.4 Payoff matrix for case ii) . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 31
Fig. 4.5 Payoff matrix for case v) . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 31
Fig. 4.6 Mixed strategies for the Avranches Gap Situation.. . . . . . . . . . . . . . . . . . . . 34
Fig. 4.7 Mixed strategies for ordinal payoffs, case v. .. . . . .. . . . . . . . . . . . . . . . . . . . . 35
Fig. 5.1 Payoff matrix for the mining model . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 40
Fig. 5.2 Game tree for the mining model . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 41
Fig. 5.3 Payoff matrix for A2/AD . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 45
Fig. 5.4 Game tree for A2/AD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 46
Fig. 5.5 Payoff matrix for A2/AD with opportunity costs . . . . . . . . . . . . . . . . . . . . . 47
Fig. 7.1 Security in Iraq, January 2004–March 2010 .. . . . .. . . . . . . . . . . . . . . . . . . . . 60
Fig. 7.2 Payoff matrix for “The Surge” . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 64
Fig. 7.3 Payoff matrix for “Clear, Hold and Build” . . . . . . .. . . . . . . . . . . . . . . . . . . . . 69
Fig. 7.4 Payoff matrix for “Clear, Hold and Build” C “The Surge” . . . . . . . . . . 72
Fig. 7.5 Payoff matrix for the enemy-centric surge . . . . . . .. . . . . . . . . . . . . . . . . . . . . 73
Fig. 7.6 Payoff matrix for the enemy-centric surge with asymmetric
information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 75
Fig. 8.1 Payoff matrix for ISAF operations . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 84
Fig. 8.2 Payoff matrix for ISAF operations with asymmetric information . . . . 86

xi
List of Tables

Table 7.1 “The Surge”. Expected outcomes .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 65

Table 7.2 Actions in “Clear, Hold and Build”. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 67
Table 7.3 “Clear, Hold and Build”. Expected outcomes .. .. . . . . . . . . . . . . . . . . . . . . 70
Table 7.4 COIN strategies ranking . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 77
Table 7.5 Summary of findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 79

xiii
Chapter 1
Introduction

Military literature often not just suggests a picture of war characterized by asym-
metric information, thereby evoking what Clausewitz called the “fog of war”,
but quite often implicitly assumes that the force that is able to command more
information than its enemy should be more likely to carry the victory. Until recently,
it has dominated—and effectively still does—all military doctrinal teaching at least
in the Western sphere. The most recent example is the Network-Centric Warfare
concept which has become something like a departure point for, if not the Holy Grail
of, almost every doctrine of every branch of the US military since the beginning
of the 1990s. By putting the emphasis almost exclusively on (the compatibility of)
communication networks and using catchwords such as, e.g., “red-force tracking” or
“blue-force tracking” it suggests that warfighting is about little else than acquiring
information. Translated into game theory, it says that whenever war resembles a
sequential game, there would be a second-mover advantage, implying that it should
be every force’s primary objective to turn a simultaneous-decision-making military
encounter into a sequential one by, e.g., trying to find out about the enemy’s
objectives, his positions or his battle plan. Or, to put it in other words, in war,
information should always pay.
The purpose of this thesis is to scrutinize this argument by paying closer attention
to the underlying (implicit) assumptions and to use both classical war scenarios as
well as battlefield situations found in modern, and in particular irregular, wars to
illustrate the findings.
Chapter 2 reflects on both the role of best-response strategies and their usefulness
for modeling conflicts. It discusses, inter alia, the assumption of rationality for
explaining human behavior, and the restrictions one has to accept when using best
response strategies. Chapter 3 introduces conceptual key issues necessary to discuss
decision-making in military affairs. It provides an historical review of the conduct of
war that would explain the actual organization of conflicts. In particular, it is impor-
tant to understand that the term “strategy”, as used by the military and referring to
both the objectives of a nation as well as ways how these could be best achieved

© Springer International Publishing AG 2017 1

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_1
2 1 Introduction

alike, is completely different from its game-theoretic meaning. In particular, it is

pointed out that the models discussed in this thesis all refer to decision-making
on at most the operational level. Chapter 4 describes the relationship between the
imbalance of power and the strategies at hand to explain outcomes by analyzing
asymmetric conflicts as zero-sum games. In addition, it analyzes the extent to which
zero-sum games can give rise to second mover advantages. Chapter 5 introduces
the concept of a first mover in contrast to the military conception of moving first.
The sequential character of Anti-Access/Area-Denial (A2/AD) strategies serves to
analyze the role of being first-mover in non-zero-sum games. Chapter 6 focuses
on the concept of “Irregular Warfare” and the new challenges unconventional wars
pose to the design of operations. Chapters 7 and 8, taking account of the “Irregular-
Warfare”-nature of the payoffs in the wars of Iraq (“Operation Iraqi Freedom”) and
Afghanistan (“Operation Enduring Freedom”), discuss the role information has in
counterinsurgency operations. Chapter 7 offers a game-theoretic analysis of the two
major features of the Iraq war, one being the US decision to send additional combat
brigades to Iraq in 2007, and the other the introduction of the new combined US
Army/Marine Corps Doctrine in Counterinsurgency that turned US strategy from
an enemy-centric into a population-centric approach. The penultimate Chapter is
based on a joint work with Jörg Schimmelpfennig. It looks into the existence of
strategic moves for routine tactical operations undertaken against the Taliban during
Operation Enduring Freedom, where danger close situations could not always count
on the kind of close air support that would have been needed as the US Army did not
have a sufficient number of attack helicopters to guarantee them being available for
each and every mission. The final chapter provides a short summary of the findings
of this thesis.
Chapter 2
Game Theory and the Decision-Making Process
in Military Affairs

2.1 Game Theory: A Theory of Interdependent Decision

Assuming that any decision-maker will maximize the expected value of some
utility function, we will find different methods to decide which action among all
possible alternatives could be the most appropriate according to our interests. If
we use, for example, a multi-criteria decision-making approach we would conduct
an analysis arranging all relevant factors in a hierarchical structure. With a cost-
benefit analysis our focus will be directed to assess the strengths and weaknesses
of all possible options to determine the decision that gives us the highest possible
net payoff. There are other situations where the outcome will not only depend on
one’s own decision but also on that of other actors, and vice versa. These would
be the so-called interdependent decisions. In such cases we might be interested
in gathering information other than the one needed for a cost-benefit or a multi-
criteria analysis, like information about the intentions of other decision-makers who
are involved. Game theory represents a part of decision theory, where two or more
decision-makers are involved in the result. To carry out these analyses we assume
decision-makers to be intelligent and rational. The meaning of being intelligent in
the context of game theory refers to the assumption that each player will not only
know his possible payoffs and strategies but also his enemy’s. Therefore, he will
be able to make any inference about the game that any other external observer
sharing the same information would be able to make, too.1 Game theory aims to
analyze situations of conflict and cooperation by means of mathematical models.
The resulting models should provide guidance for either player when having to
choose a strategy in order to achieve a good or possibly the best outcome.

1
Obviously we do not observe this in all real-life situations. For more about theoretic foundations
about decision-making see, e.g., Meyerson (1991).

© Springer International Publishing AG 2017 3

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_2
4 2 Game Theory and the Decision-Making Process in Military Affairs

The nature of conflicts and thus the interaction and interdependence between
adversaries provide a prime example for interdependent decision-making. It should
not come as a surprise that one of the vintage works of applied game theory
and strategic behavior analysis, i.e. Schelling’s The Strategy of Conflict (1960)
originated right here. We observe how games such as the Prisoner’s Dilemma and
Chicken have been used in the context of international politics to model arms races,
or disarmament, for that matter, and to explain how a crisis can be caused by, e.g.,
coercion (see, e.g., Snyder 1971). Simple analysis can also be applied to estimate
the impact third parties might have on a conflict. Terris and Maoz (2005) show,
for example, how mediation can be used to raise cooperation as it can transform
Prisoner’s Dilemma situations into coordination games. Hence, we will find in the
literature of conflicts, situations where, even though the point of departure follows a
Prisoner’s Dilemma pattern, they will not adjust to non-cooperative models. This
occurs not only when we include third parties in our models but also in cases
where players value future outcomes sufficiently high so that iteration would induce
cooperation.
Many of the questions dealing with the analysis of conflicts are related to the
determination of the causes behind conflicts, investigating, e.g., the circumstances
responsible for making a conflict more likely to break out or the variables expected
to affect the final outcomes. The bargaining theory provides a good framework for
exploring these questions. Schelling (1960) was the first to consider conflicts as
bargaining situations. Since then, many scholars have continued to build a literature
on conflicts using bargaining models in order to analyze, e.g., the relationship
between the distribution of power and the likelihood of war, like Fearon (1995),
Powell (2002), Wagner (2000), and Chadefaux (2011), to explain the conditions for
the termination of a war, like Wittman (1979), or, like Kydd (2010), to concentrate
on the role of mediation for the prevention and resolution of conflicts.
There are also plenty of models whose analysis is focused on the decision of
whether to go to war. This decision is strongly affected by the uncertainty existing
over the costs of fighting. Powell (2004), e.g., considers war as a learning process
about the distribution of power between the adversaries.
In interdependent decisions we find two types of relevant information; on the one
hand, we look at the preferences and possible reactions of all players, i.e., broadly
speaking, the rules of the game. On the other hand, though, this does not necessarily
include information on the moves actually chosen by the other player(s) which, in
case of a simultaneous game, would only be revealed once the game has come to an
end, or, in case of a sequential game, would be revealed in turns while the game is
still going on. In the latter case the decisions of the second player would obviously
be conditioned by the first player’s strategy.
When using the Nash equilibrium as a solution concept for non-cooperative
games, we are looking at strategies as best responses, whether or not the game is
a sequential one. In case of a sequential game, though, we can, provided we know
about other players’ intentions, rather consider playing a strategy based on what we
think or know the other player might be doing. For instance, we may reconsider to
choose the strategy associated to the Nash equilibrium if we expect other players
2.2 Some Comments on Using Game Theory to Model Tactics 5

not to be playing their Nash equilibrium strategies. Even if we expect other players
to do so, in games where we have multiple Nash equilibria in pure strategies we
still do not have any method to predict which of the equilibria in pure strategies is
more likely to happen2: in that case any additional information about other players’
intentions could be useful. It thus seems to be worthwhile to collect information
about other players’ intentions in order to improve our own payoffs. If we are able to
anticipate an enemy’s moves we can turn a simultaneous game into a sequential one
in which we will become a second mover. Nevertheless, guessing right or achieving
relevant information to make predictions about an enemy is usually linked to costs.
Thus, collecting information would only lead to higher payoffs if there exists a
second-mover advantage. Furthermore, the costs of collecting information will also
affect the final outcomes, and should therefore be considered when evaluating the
sequential version of the game.
In any military context, there is a whole range of information, such as weather
forecasts or, knowing the social and economic impact of any battle, that can turn out
to be crucial. The most relevant information, though, is the one related to an enemy’s
capabilities. In a military context it is commonly assumed that every force has a high
willingness to pay for information, particularly if it is this information that can be
used to guess the enemy’s next moves. Information is always assumed to be a useful
tool for commanders while designing operations. This would mean that intuitively,
expressed in game theoretic terms, we will expect conflicts where every force would
have an incentive to turn the simultaneous game into a sequential one, where either
each force would enjoy a second mover advantage or will bring an outcome at least
as good as the one in the simultaneous version. It is but equivalent to assuming that
commanders will always enjoy a second-mover advantage, regardless of the specific
battle situation. It is one of the main objectives of this study to highlight the extent
to which one can consider this (implicit) view to hold true.

2.2 Some Comments on Using Game Theory to Model

Tactics

The extent, or degree, of human rationality when it comes to making decisions

has for long been one of the focal discussion points in the social sciences.
Economists, philosophers and sociologists have been striving to provide a number
of theories looking for a better explanation of human behavior than the classical
homo oeconomicus. Accordingly, one of the biggest representations against game
theory, in particular when it comes to modeling social systems—and war would

2
In coordination games focal points might exist—if there is some property all players can
recognize—making one of the equilibria to stand out from the others. Each player would expect
the other to recognize the attribute that makes this equilibrium to stand out and therefore play it
(Schelling 1960, p. 57).
6 2 Game Theory and the Decision-Making Process in Military Affairs

be just one example—, has been the criticism of the assumption of players being
rational individuals. In a mitigating way, some authors argue that “rationality” is
itself being affected by other factors, and therefore one should focus on these factors
when wanting to explain real players’ decisions.
Standard economic theory of course assumes rationality and thus, agents who
maximize their expected utility. Nevertheless, the idea of rationalism has been
continuously contested. Starting in the eighteenth century, the Scottish philosopher
David Hume challenged the idea of rationality. In his Treatise of Human Nature
Hume assumed that people reason by analogies and therefore decisions would
result from induction and not from causal reasoning (Hume 1985, pp. 117–230).
He suggested that, following the principle that nature will continue to behave
in the future as it has in the past, individuals would decide relying on previous
experiences. The history of similar past events would lead them to build expectations
on what is more likely to happen in the future. However, according to Hume,
inductive reasoning leads to a fallacy, since for any cause there are multiple effects
conceivable and expecting that an event is more or less likely to happen in the future
because it happened in the past would be irrational. In addition, Hume’s system of
preferences based on individuals’ desire and aversion would require to be constantly
revised, since emotions are not constant but subject to continuous change (see,
e.g., Gilboa and Schmeider 1995). But it was only in the 1950s when the standard
assumption of rationality in social sciences started to be questioned. Simon (1957)
coined the term “bounded rationality” to indicate that assuming complete rationality
was rather unrealistic.3
Rationality is also contested in the field of psychology, giving rise to a crit-
ical positioning towards the use of game theory for describing social behavior.
Von Bertalanffy (1968) considers game theory as a method that would only be
reliable when used to explain survivor patterns and animal evolution. The distinction
between individuals’ decisions and animal behavior relies on the fact that animals’
actions are not the result of decision-making but rather the outcome resulting from
following a genetic behavior pattern; hence there is no possible deviation from the
expected behaviour.4 Individuals on the other hand are way too sensible to other
factors that might influence their decision, leading to an alteration and, eventually,
a different final outcome of the game. It is not just rationality that has become a
focus when discussing the practicality of game theory as a means to predict players’
behavior. Taking, e.g., the idea of the Nash equilibrium as a solution concept for
non-cooperative games, we have to assume that players have to be certain that
the other player(s) will use the Nash solution concept, too. Otherwise they might
have an incentive to deviate from playing Nash as well. However, the conditions

3
According to Simon, emotional impulses often override rational deliberations. One could
think individuals would rather satisfy than optimize. Selten (1998b) introduced the “Aspiration,
Adaptation Theory” where individuals would make choices in order to satisfy some aspiration.
Selten (1998a) also found experimental evidence pointing to a substantial deviation from Bayesian
rationality.
4
Biology has a record of a great influence of using concepts of evolutionary games.
2.2 Some Comments on Using Game Theory to Model Tactics 7

under which two players play strategies leading to a Nash equilibrium might not be
feasible.
The first problem with the Nash equilibrium as a solution concept arises in cases
when we have multiple equilibria because we cannot predict which equilibrium is
more likely to happen. We may consider that a solution by multiple equilibria could
be given by negotiation between players, but this does not always solve the problem,
as there are many cases where players have an incentive not to keep the negotiated
agreements.5 There are also cases where one of the Nash equilibria is “focal” and
players will be expected to recognize it and therefore play the strategy linked to the
focal Nash equilibrium (Schelling 1960). However, as focal points vary depending
on personal background and culture, to recognize them would require players to
share some common knowledge. But even in games with a single Nash equilibrium,
the Nash equilibrium might not be the solution expected.
Bernheim (1984) and Pearce (1984) investigated the degree to which the Nash
equilibrium should be expected to be an appropriate solution concept in the first
place. Bernheim (1984) considered that there is no guarantee that players may
not make mistakes after all and coined the term “rationalizability“ to express a
descriptive term more flexible than the Nash solution. Pearce (1984) focused on
situations where players were not able to communicate with each other before or
during the game. The strategy a player would choose would be the best response
to his conjecture about other’s strategies, not to his actual strategies. Pearce (1984)
assumed that although players have a common knowledge of the structure of the
game,6 they do not have an objective probability distribution over other player’s
choice of strategies and would therefore have to form a subjective prior probability
distribution that does not contradict the actual information they might have. Then
each player will maximize his expected utility based on this subjective distribution.
Selten (1975) introduced the idea of imperfect equilibrium indicating that complete
rationality was only a limiting case of incomplete rationality. Another example of
the restrictions implied by a Nash equilibrium solution is given by Fudenberg and
Levine (1993), who find an alternative equilibrium by weakening the assumption
of having players with perfect beliefs about the enemy’s strategies. Instead, they
assume players only have correct beliefs along the equilibrium path. It means,
players are allowed to have incorrect beliefs in contingencies that will not arise.
They find that the resulting equilibrium, named self-confirming equilibrium, might
differ from the Nash outcomes. Then, rationality alone does not imply that players
always follow strategies leading to the Nash equilibrium, but that players are

5
We find this possibility in non-cooperative games when players have the opportunity to commu-
nicate before they play. In such cases they can reach an agreement or coordinate their strategies in
a way it benefits both. Such scenarios would lead to so-called “coalition-proof Nash equilibria”.
See Bernheim et al. (1987).
6
Common knowledge is referred here in the sense of Aumann (1976) “Two people, 1 and 2, are
said to have common knowledge of an event E if both know it, 1 knows that 2 knows it, 2 knows
that 1 knows it, 1 knows that 2 knows that 1 knows it, and so on” (Aumann 1976, p. 1236).
8 2 Game Theory and the Decision-Making Process in Military Affairs

affected by the learning process they go through when they do not have perfect
information.7
We also find factors that affect the utility functions in unexpected ways, resulting,
e.g., in players willing to decrease their payoffs in order to see the other player’s
payoffs increased; or the opposite, that players are willing to give up part of their
payoffs only to punish other players. This behavior is observable when players are
affected by feelings like sympathy or hostility towards the other player (see, e.g.,
Rabin 1993 and Fehr and Schmidt 2000). We find that players tend to include in
their decisions deliberations on what they consider fair or any feelings about having
been treated unfairly by other players and are therefore likely to care not exclusively
about material payoffs but also about fairness, equity and reciprocity. This pattern of
behavior has been widely observed in the ultimatum game where we find two players
having to decide on how split a given amount of money. The first player (known as
the proposer) proposes a partition and the second player (known as the responder)
has two options: to accept the partition proposal in which case both players would
get the share proposed by the first player or to reject it. By rejecting it, none of
the players would get anything.8 The results shown by experimental game theory
contradict the rationality assumption but they are rather intuitive nonetheless: first,
even in scenarios when the game is played just once between any two randomly
selected players, partitions allocating a smaller but still non-zero share to the second
player have an increasing higher probability of being rejected, and, second, the
rejection patterns depend on the countries where the game is played.9
Obviously, though, laboratory results should, as pointed out by Levitt and List
(2007), always be considered with a grain of salt, as the underlying behavior cannot
necessarily be assumed to be the same as if the game was played in real life
and, thus, with not just a token couple of dollars to be distributed. But even if
those results are biased, we still can agree on the role of information. If we take
for example the ultimatum game, having information about other players’ cultural
background and their concept of fairness can be an asset to better anticipate possible
reactions to one’s own actions, and could therefore enable a player to adjust his
beliefs accordingly. We thus find an incentive to collect information to identify the
type of player one is dealing with, in order to improve on one’s best responses and
subsequently achieve a higher payoff.
Boulding (1962) adds two other aspects to be taken into account when modeling
social sciences. First, individuals are moved by distant goals and not only by

7
An altogether different approach would of course be to change the design of a game in such a way
that players would no longer have any incentive whatsoever to deviate from their respective Nash
equilibrium strategies as authors like, e.g., Maskin (2008) have suggested.
8
Widely in Thaler (1988).
9
Hoffman et al. (1994) found evidence that showed that in addition to fairness there were other
factors affecting the behavior of players in the ultimatum and the dictator game. First movers
would act differently if their role is not assigned randomly but earned. And second, they would act
affected by the social concern about what others may think, and therefore establish a relationship
between anonymity and the level of fairness with which the payoffs were distributed.
2.2 Some Comments on Using Game Theory to Model Tactics 9

immediate pressures. In defense, for instance, when considering the range of options
available to a commander, we may identify situations where the commander may
even consider intentionally losing a battle, as long as it might help to win the
campaign. Therefore, decisions might not be rational if they are looked upon from
a narrow viewpoint but may be perfectly rational regarded in a broader context.
The more distant goals, though, are also susceptible of change as information
arises, implying that with hindsight a decision may appear to have been completely
“irrational” irrespective of the respective perspective.
Boulding (1962) applies mechanical dynamics in social systems to explain
conflicts. In particular, the path along which players move can be illustrated by
the use of differential equations. At the same time, though, he highlights the main
caveat to applying dynamic systems to social systems: Social systems, contrary to
most physical systems, are not dynamically stable on the small scale. An illustration
of this could be the Cobweb theorem, where expectations and the time lag between
the decisions for demand and supply are used to explain price fluctuations (Boulding
1962, p. 22).
Second, Boulding (1962) also refers to the uncertainty involved in expecting
other players’ moves. In particular, any supposed “guarantees” about another player
adhering to a certain strategy are not tenable when it comes to deducing the expected
outcomes. By assuming players in a game to be intelligent, we effectively assume
that they have perfect knowledge about the outcomes resulting from all possible
strategy combinations. Boulding (1962, p. 57) argues, though, that a player can
neither be sure about knowing the strategies of other players nor the payoffs
associated to each possible combination of strategies.
On top of the challenges of dealing with distant goals and assuming rationality,
when using game theoretic models we have yet to adopt another restriction: When
designing game theoretical models for making predictions, we usually need to
reduce the complexity of reality to a conceptual skeleton, and by doing so we have
to accept a certain risk of oversimplification. Such a simplification may in turn give
rise to the not even remote possibility of facing completely unexpected events and,
thus, outcomes.
Despite all these qualifications, though, we find several authors employing
game theory to model conflict situations (see, e.g., Rapoport 1974). Expected
utilities are often used to explain conflicts’ outcomes. A good example of this is
Bueno de Mesquita’s theory about the conditions necessary for war.10 In The War
Trap, Bueno de Mesquita (1981) assumes that the people responsible for making
decisions in conflicts are rational and able to estimate the expected utilities linked
to each strategy. In addition, his models also foresee the fact that decision-makers
will not necessary be homogeneous and will have different reactions to risk and
uncertainty.11 In “War and Reason” Bueno de Mesquita and Lalman (1992) rely

10
Bueno de Mesquita’s theories in The War Trap refer to interstate conflicts.
11
The model also includes expected utility functions for third countries considering whether to
enter into a military alliance or not (Bueno de Mesquita 1981, pp. 59–91).
10 2 Game Theory and the Decision-Making Process in Military Affairs

on the application of rationality in order to explain the circumstances that would

cause a war to break out. Their model explains the case of two countries—one of
them considered the initiator and the other one the target—deciding on whether to
make some kind of demands and use force in order to push them through, resulting
in a sequential move game where under perfect information a backward induction
analysis delivers the expected result.12
In summary, to assume rationality and perfect information in social sciences
seems often difficult to justify. Still, applications of game theory to military conflicts
have, at least occasionally, led to additional insights providing better explanations
for the observed behavior of warring parties. It is not just historical conflicts, though,
this study is interested in, it is rather operational planning as a whole which is looked
at from a game-theoretic perspective, in order to offer guidance for developing
strategies for future, and different kinds of, military conflicts.

12
In addition, Bueno de Mesquita and Lalman (1992) provide empirical evidence of their model
analyzing 707 cases from 1876 to 1970 that support their results.
Chapter 3
Strategies and Tactics

3.1 A Brief History of Military Thought

The words strategy and tactic share an etymological and historical origin. While the
word strategy is etymologically formed by the Greek stratos (army) and agein (to
guide), meaning the art of leading military operations, the word tactic comes from
the Greek taktike (tekhne), meaning the art of arrangement. It was the Byzantine
emperor Leo VI who around 900 first used the word strategía to express how
to move armed forces, and who released the military treatise “Tactica” collecting
different issues in the conduct of war, from cavalry and infantry formations to naval
operations. In the Western world, though, it took until 1771 when Paul Gédéon Joly
de Maïzeroy translated the works of Leo VI into French that the term stratégie was
first coined, stressing the difference between the specific art of how a commander
has to successfully move his subordinates and other aspects of a military campaigns
such as tactics, logistics or building fortifications (see, e.g., Heuser 2010, p. 3).
Interestingly, there is a manual dealing with strategy that has survived from
Roman times and was frequently read and studied until well into the middle ages:
The “Epitoma rei militaris” by Flavius Vegetius Renatus dates from the late fourth
century and is a compilation from other sources on Roman military institutions. It
not only highlights, e.g., the strategic value of information—“No plans are better
than those you carry out without the enemy’s knowledge in advance” (Vegetius
1993, p. 116)—but can be understood as a document with a profound political
significance and a clear intention to reform both the institutions as well as the
underlying strategic thinking of the Roman Empire, in particular by linking the
future of the Roman legion, as an institution, to that of the whole Roman empire
(Vegetius 1993, pp. xxvi–xxviii). Vegetius was convinced that one of the reasons for
the deterioration of the latter was the lack of competent soldiers. The long periods of
peace produced a change in the society that provided better incentives to enter into
civil occupations rather than embarking on a military career (Vegetius 1993, p. 6).
In combination with an increase in governmental corruption and the almost

© Springer International Publishing AG 2017 11

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_3
12 3 Strategies and Tactics

habit-forming way of getting rid of emperors by means of assassination, thus leading

to an ever more blurred division between the two spheres, the military became too
corrupted to be able to successfully fend off the increasing numbers of barbarian
incursions into the empire and, eventually, Rome itself.
Around 300 years before Clausewitz defined war as the continuation of policy by
other means, Machiavelli had already highlighted the central role of military power
in political life. As Machiavelli himself put it in two of his most important works
“The Prince” and “Discorsi”: “There cannot be good laws where there are not good
arms, and where there are good arms there must be good laws” (Machiavelli 1984,
p. 336).
Machiavelli was convinced of the existence of a general norm valid for military
organization of all states and times. He did not believe wars were a matter of chance
mostly decided by the goddess Fortune—as it had been widely believed in the
Middle Ages—, but by the supremacy of reason. If military organizations followed
what was prescribed by reason, then it should be possible to reduce the influence of
chance and assure success (Gilbert 1986, pp. 23–24).
On the other hand, the new laws of war Machiavelli intended to introduce in
his “Art of War” were nothing less than the old laws of the Roman military order.
While Machiavelli lacked any military training himself, he was nevertheless devoted
to regenerate military structures, thereby quoting Rom as the perfect example of
turning a city-republic into a world power. Even though his critics turned on his
lack of knowledge in military affairs and sometimes described his “Art of War” as
a simple transcription from the works of Vegetius, Frontinus or Polybus, the value
of adapting Vegetius’ ideas to the needs of the time was acknowledged nevertheless
(Gilbert 1986, p. 22).
The transition to the feudal system gave rise to flowering city-states that could not
afford to set business aside to supply soldiers to civic militias. Moreover, military
skills and tactics were affected by the demand for external armies, tending towards
specialization: German mercenaries were sought after for their effective riding
formation, Hungarians because of their skills as lightly armed riders and Italians as
foot soldiers. The composition of these professional foreign forces changed military
tactics that had traditionally relied on foot soldiers only, to tactics attributing the
leading role to cavalry (Brauer and Tuyll 2008, p. 87).
Machiavelli was very concerned about the ruling mercenary system prevailing in
Italian city-states like Venice, Florence and Genoa which, during the twelfth and
fourteenth century, had relied on mercenary forces, the so-called condottieri, to
defend themselves against foreign attacks. The condotta was the contract signed
between city leaders and mercenaries, where the terms and conditions of their
military service were stipulated. Machiavelli’s critique of the condottiere system
was based on, what today would be called moral hazard and information asymmetry.
Since the condottieri basically saw war as a business, it should not, according to
Machiavelli, have come as a surprise that they tended to last ever longer. There
was a clear incentive to extend the duration of conflicts over time because once the
conflict was over, all the condottieri involved in it would lose their occupation: war
became something of a golden goose that must not be killed.
3.1 A Brief History of Military Thought 13

In addition, mercenary forces became ever more reluctant to take risks or expose
themselves to the danger of life and limb. Machiavelli observed that the number
of casualties in conflicts involving the condottieri were not consistent with the
casualties one would expect for the kind of battles fought at that time. Machiavelli
argued that, e.g., in the battle of Anghiari “lasting from the 20th to the 24th hour,
only one man was killed and he was not wounded or struck down by a valiant
blow but fell from his horse and was trampled to death”, and consequently: “That
cannot be called war where men do not kill each other” (Machiavelli 1989, p. 1281).
The lack of information put a great strain on city leaders who were incapable of
making objective assessments of the severity of the conflicts they were facing. The
ensuing principal agent problems between city leaders and the condottieri became
insurmountable and effectively caused the end of the mercenary period, at least
in the form of city-state defense. The mercenary system lasted until the middle
of the fifteenth century, when Europe started a new period of militarization and
castellation.
The invention of gunpowder and the rise of nation-states in the seventeenth and
eighteenth century changed the military landscape. Since the building of Jericho—
the oldest fortified town known in the world—the basic defensive features such
as walls, moats and towers had, over the next 8000 years, experienced little
change. Gunpowder, though, not only rendered many of the traditional fortifications
obsolete, but due to the exponential cost of the new technology, private warfare
became financially impossible: It was only the state, being able to raise taxes on a
large scale, that could provide financing for wars, and as economies grew, so did
the budget for war. The scientific revolution and the organization of states changed
warfare from castle-based to army-based.
In the eighteenth century the regiment was fully established in the European
military landscape.1 Charles VII’s compagnies d’ordonnance arose in the middle
of the fifteen century as mercenary companies with a uniform composition and
the officially recognition as servants of the monarchy. The superiority of feudal
cavalry over the infantry of the first colonelcies of infantry was undermined by
the entry of the gunpowder. Although at the beginning in most European armies
these regiments belonged to the regimental colonels, they rapidly became permanent
royal institutions, with officers being recruited from the aristocratic families (see e.g.
Brauer and Tuyll 2008; Keegan 1993).
In 1792 Carl Philipp Gottlieb von Clausewitz, son of a Prussian tax receiver,
joined one the regiment “Prinz Ferdinand” at the age of twelve. He eventually
became the youngest Prussian general and one of the most prolific writers on
military theory. His works—based on his military experience and embracing ten
volumes, seven of which being dedicated to military history and the first three
to military theory—have since become part of the reading lists of most military

1
The word regiment itself points out the relationship of this formation and the government. It
comes etymologically from the latin “regere” (govern) which shares the same root with “rex”
(king), meaning: the way of governing.
14 3 Strategies and Tactics

academies all over the world. It were precisely the first three volumes—though
unfinished and published by his widow—called “On war” that assured Clausewitz
posthumous fame (Strachan 2007, p. 17). Clausewitz denied the existence of any
positive theory of war because theory could hardly fit reality. Each single campaign
would take place in a set of conditions unlikely to be found twice. Thus, theory
would not be of much use if compared with experience and talent. For Clausewitz,
the higher the military rank taking a decision, the less any theory could be of any use.
At the top of the decision chain, i.e. at the level of the general staff and commander-
in-chief, all decisions would depend solely on his genius and his talent. Clausewitz
coined the phrase “fog of war” to summarize the uncertainty prevailing in all levels
of warfare. The concept of “fog of war” still continues influencing to this day the
doctrines on how to fight war. The ability of a commander is all the more significant
when it comes to facing a lack of information or the fact that the information one can
obtain is dubious and often erroneous. Thus, according to Clausewitz, in the fog of
war the question of whether balance is tipped one way or the other would critically
depend on the respective commanders’ capabilities of managing decisively. Besides
leadership, Clausewitz identified another two critical features in the conduct of war:
He argued in favor of the strategic value of defense over that of attack, with defense
constituting a superior form of warfare, and he introduced the concept of the decisive
battle.
The study of warfare was traditionally divided into either strategies or tactics.
The term tactic was used to describe any action within the range of enemy fire while
strategy was reserved to all military activity beyond the tactical range (Daase 2007).
However, Clausewitz’s definition of strategy was characterized by his concept of
war being the continuation of policy by other means. According to him, it was the
purpose of war that gives form to strategies:
“Strategy is the use of the engagement for the purpose of the war. The strategist must
therefore define an aim for the entire operational side of the war that will be in accordance
with its purpose. In other words, he will draft the plan of the war, and the aim will determine
the series of actions intended to achieve it: he will, in fact, shape the individual campaigns
and, within these, decide on the individual engagements” (von Clausewitz 1976, p. 177).

As both the industrial revolution as well the French revolution had contributed
to the levels of complexity involved in the conduct of war, two further concepts,
within the two broad categories “strategies” and “tactics”, were offered: Grand
Strategy would be used for the setting of national objectives, while Grand Tactics
would involve tactics related to larger forces that could be active in multiple
battlefields (Doane 2015). The Swiss officer Antoine-Henry Jomini, contemporary
of Clausewitz, considered that the concept of Grand Tactics explained the difference
between planning and implementing strategies on the ground:
“Strategy is the art of making war upon the map, and comprehends the whole theater of war.
Grand Tactics is the art of posting troops upon the battlefield according to the accidents
of the ground, of bringing them into action, and the art of fighting upon the ground, in
contradiction to planning upon a map. Its operations may extend over a field of ten or
twelve miles in extent. Logistics comprises the means and arrangements which work out
the plans of strategy and tactics. Strategy decides where to act; logistics brings the troops to
3.1 A Brief History of Military Thought 15

this point; grand tactics decides the manner of execution and the employment of the troops”
(Jomini 1862, p. 69).

In the context of Grand Tactics information played a key role. In order to secure
the optimal use of the resources available, the commander had to assess the options
he had in the battlefield to identify some “objective points” which were supposed
to be the ones with a greater probability of success. Jomini distinguished between
two types of objective points, points of maneuver, referring to and dealing with the
positioning of enemy forces, and geographical objective points, like geographical
features, that could either be natural like, e.g., the line of a river, or man-made, such
as, e.g., a fortification (Jomini 1862, p. 88). Jomini considered the conduct of war
as an art the strategies of which would be constrained by fixed laws.
Following WWI, a growing number of military theorists attempted to further
clarify the terms strategy and tactics as their use was often discretionary. The
English soldier and military theorist Sir Basil Henry Liddell Hart supported the
idea of strategy as a concept to be understood only in conjunction with policy. He
used the above-mentioned concept of a Grand Strategy to illustrate the relationship
between policy and strategy:
“As tactics is an application of strategy on a lower plane, so strategy is an application on a
lower plane of ‘Grand Strategy’. . . .While practically synonymous with the policy which
guides the conduct of war, as distinct from the more fundamental policy which should
govern its objective, the term ‘Grand Strategy’ serves to bring out the sense of ‘policy
in execution.’ For the role of Grand Strategy—higher strategy—is to coordinate all the
resources of a nation, or a band of nations, towards the attainment of the political object of
the war—the goal defined by fundamental policy” (Hart 1991, p. 335).

Furthermore, Liddell Hart introduced of the concept an “indirect approach”

of strategy. It is related to Jomini’s strategic points in the sense of rather than
concentrating on dealing with one’s enemy’s strengths one should instead explore
his weaknesses. For the French general André Beaufre, contemporary of Liddell
Hart, strategy was defined by two general principles: freedom of action and
economy of force. He proposed a new concept of strategy, the “action strategy”,—
which consisted in making someone do something—as a counterpart of deterrence.
His most singular contribution was a universal explanation of strategy based on one
formula with variables such as material force at disposal or psychological factors
(Bartholomees 2010, p. 24).
Today, the Department of Defense Dictionary of Military and Associated Terms
defines strategy as ”A prudent idea or set of ideas for employing the instruments of
national power in a synchronized and integrated fashion to achieve theater, national,
and/or multinational objectives” (Joint Chiefs of Staff 2015, p. 229) and tactics as:
“The employment and ordered arrangement of forces in relation to each other” (Joint
Chiefs of Staff 2015, p. 237). The definitions suggest a concept of strategy that holds
the creative part in the design of operations, while tactics deals with supposedly
more mundane tasks such as execution and coordination.
16 3 Strategies and Tactics

3.2 Doctrine and the Organizational Levels of Conflicts

Throughout the evolution of conflicts, we find no common underlying theory for

all levels of military decision taking. “Doctrine” serves as a reference to standard
warlike confrontations. It was not until the Cold War, though, that “Doctrine” was
standardized by the NATO Glossary of Terms and Definitions as “fundamental
principles by which the military forces guide their actions in support of objectives.
It is authoritative but requires judgement in application” (NATO Standardization
Agency 2013, p. 2-D-9).
Historically, doctrine has been used for the tactical level. Its use has mainly
focused in providing cohesion in military decisions because it facilitates synergetic
effects inside the forces. By providing a catalogue of maneuvers, weapon choices or
types of troops to be used in certain textbook-like situations, they represent a kind
of default approach for situations similar to that described in the manual.
However, doctrine can also make a force become more predictable to its enemy
since it could enable him to predict the kind of actions he can expect in the
battle. For example, the latter “rules of engagement” for coalition forces during
the Afghanistan war enabled the Taliban to enjoy a higher movement of action as
they knew that they could avoid being attacked as long as they withheld their fire.
Rather than dealing with the tactical level alone, though, the 1993 edition of
the US Army’s keystone war fighting manual, Field Manual 100-5 “Operations”,
distinguishes three levels of war: the strategic, the operational and the tactical
level. Each one refers to different scopes of the decision-making process. While
the strategic level deals with decisions affecting the nation, the operational and the
tactical level contribute to military success through the design and execution of
the force’s deployment. The operational level is then considered to link strategic
objectives with the execution of battles and engagements through the design,
organization and conduct of campaigns (Department of the Army 1993, p. 1–3).
Decisions at the strategic level have a political dimension, implying that the
impact of a conflict has to incorporate political costs as well. It is at the strategic level
where we find the link between politics and the use of force. The Romans considered
war to be an option only if the fetialis, a group of special priests, had decided that the
enemy—a foreign nation—had violated its duties towards the Romans. Approval by
the collegium fetialium thus assured Roman soldiers that, once fighting had started,
they would be supported by the gods. Interestingly, the very same idea led to the
regulations on the use of force as a means of war, i.e. the justifiableness of war,
being incorporated into Roman law (Nussbaum 1943, p. 454). In today’s times, it
is of course “matters of national security” that are most often called upon when
decisions to go to war, i.e. on the strategic level, are made. One only has to look at
the “War on Terror” or the fight against ISIS.
The operational level, in the same way as the concepts of Grand Strategy and
Grand Tactics did, only emerged from the challenges of a more modern warfare at
the beginning of the twentieth century, as strategies and tactics in the traditional
sense could no longer cope with the then new type of conflicts and in particular
3.3 Some War Strategies 17

battlefield situations such as trench warfare. However, some authors like, e.g.,
Luttwak (1987) support the idea of the operational level not only as a bridge between
the strategic and the tactical levels, but also as an attempt to pool the competences
needed for and characteristics exhibited by both levels of military conflict, which
military doctrine had failed to properly address until then (see, e.g., Naveh 1997).
On the tactical level, contrary to what we find in the strategic and the operational
level of war, we are able to identify a set of actions an army might consider in the
context of a battle in much more detail. The more confined the context, the easier
it should become to agree on respective objectives, and, subsequently, predict the
outcome each tactical force will pursue; provided, of course, that it is free to operate
from any operational or strategic restrictions.
While the models and scenarios to be discussed below can generally be placed
on the operational level, some of the decisions that are considered could well be
described as of a tactical nature, too. The “strategies” in our games, though, will
refer exclusively to the tactics of either of these two levels.

3.3 Some War Strategies

Based on the DoD definition of strategy we observe the following war-winning or

war-terminating strategies: attrition, exhaustion, shock and awe, and annihilation.2
The resulting victory through annihilation and attrition differs in that, while anni-
hilation ends with our enemy’s ability to carry on, attrition would be accomplished
once the enemy has suffered so many losses, materiel or personnel alike, that victory
has become highly unlikely or that the costs of a hypothetical victory would become
unacceptably high (Bartholomees 2010, p. 9).
The difference between attrition and exhaustion lies in the fact that, while
attrition is constrained to military operations, exhaustion works in a broader scope
by making the enemy, including both the soldiers and the civilian population, lose
its faith in the ability to win: As the latter would no longer be willing to support the
war effort, the former would become even more disincentivised to carry on fighting.
Exhaustion would require more resources than a purely military operation.
Moreover, what differentiates attrition from annihilation or shock-and-awe
strategies is that while the first looks for military exhaustion, annihilation and shock-
and-awe strategies aim at temporarily destroying the capacity of enemy soldiers to
fight as a cohesive force by, e.g., paralyzing the enemy’s command and control
systems, thus enabling one’s own troops to win a single pivotal battle or a limited
number of engagements sufficient to make the enemy surrender. An example of a

2
We should distinguish here between attrition as strategy in a conflict and the “war of attrition”
in game theory. The strategy attrition refers to a strategy within a military context while “war of
attrition” would describe the situation were two parties compete for a resource incurring in costs
over the time the contest lasts.
18 3 Strategies and Tactics

shock-and-awe strategy would be the US Blitzkrieg-style military interventions in

Kuwait and subsequently Desert Storm and then again operation Iraqi Freedom.
Both attrition and exhaustion are based on achieving a strategic advantage
that will cause the collapse of the enemy by continuously inflicting casualties.
Respective operations will therefore be based on offensive campaigns. Adapting
the offensive role allows the attacker to modify his operations regarding the
enemy’s options. The attacker has the option of stopping the attacks if the situation
unexpectedly changes and he starts taking an unacceptable number of casualties.
Furthermore, any strategy assuming an attacking role comes with a larger freedom
of action than that of the defender. Then, it allows the attacker to decide how, where,
and how hard an operation is going to be, resulting in a greater control of the
expected losses.
As both attrition and exhaustion aim to erode the enemy’s will to fight, attrition
strategies from the 1930s onwards have been characterized by an overwhelming
use of air power. Air attacks strive towards inflicting the highest possible physical
damage to—using Clausewitz’s words—break the will of the enemy.3 While they
can of course be directed exclusively against military objectives while at the same
time trying to minimize any civilian casualties—as had been the case with US Army
Air Force bombing campaigns right until early 1945—civilian casualties may, on
the other hand, at least be approved of, if not even explicitly be aimed for, as it soon
became the case with Germany and Britain alike with night bombings becoming
almost the norm from the middle of 1940 onwards (see, e.g., Dorr 2012).
Exhaustion and attrition campaigns have the sole objective of hurting the enemy;
other issues like territory captured or other measures of effectiveness will only be
regarded as secondary concerns. One classical example of attrition is the WWI battle
of Verdun in 1916. The main (and possibly only) objective of the German army in
this battle was to inflict an unacceptable amount of casualties on the enemy force in
order to make him capitulate.
Considering exhaustion and attrition as strategies looking at weakening the
enemy’s will power, it might seem reasonable for weaker forces, in conflicts with
large asymmetries in resources distribution, to try to damage the stronger force’s
political will rather than to target unattainable military objectives. In line with this
idea, Bartholomees (2010) argues that even if attrition favors the large force and the
attacker, it also may embody the only chance of winning for the insurgents. Once
insurgents become aware of their forlornness when it comes to winning a battle
militarily, they could resort to changing the character of the war to one of will,
which would be the only area where they hope for carrying an advantage. They
would be using another kind of attrition where the focus of the strategy would no
longer be to drain the larger force’s materiel resources but to identify ways or targets
with the highest returns on attacks in terms of morale or political damage like, e.g.,
the use of IEDs (Improvised Explosive Device) or suicide attacks. According to

3
Douhet (1998) and Mitchell (1930) introduced in the 1930’s the essential role of air power to
avoid casualties by engagement of land armies and to include civil population in the attacks.
3.4 Network-Centric Warfare 19

Bartholomees (2010) this kind of “moral attrition” against the larger force would
enable insurgencies to gain time to build forces until they generate sufficient support
in order to no longer be the underdog, eventually turning the conflict from an
insurgency into one looking more like a conventional war. The most recent example
of moral attrition can be observed in the way ISIS deals with prisoners, releasing
explicit barbarous images and videos on how they kill them—like, e.g., the release of
a video with the execution of a Jordanian pilot who was burned alive in a cage after
having being captured when his plane crashed during an airstrike—only looking for
winning support through terror.
Another example can be found in the Maoist insurgency theory, which supports
the idea of insurgents turning conflicts through attrition to more conventional wars.
Thomson (1966) in Defeating Communist Insurgency: Experiences from Malaya
and Vietnam, was one of the first authors who criticized the search-and-clear
and search-and-destroy operations in Vietnam. These missions, causing many civil
casualties and huge destruction in villages, resulted in an increasing support and
enrolment for the insurgency cause. We can observe two effects; on the one hand
attrition would result here in an increase of popular support and adhesion to the
insurgency cause, thus reducing the asymmetries between the two forces, and on
the other hand, it worked to erode the civil and political support for the larger force.
The experiences of Malaya and Vietnam required a review of the military
doctrine, not only due to the change in nature of conflicts but also due to the growing
importance of the so-called Operations Other than War (OOTW).
Finally, attrition often causes friendly casualties and it quite often takes a rather
long time before it can be concluded. Moral attrition, however, might easily turn
into political erosion. Moral attrition can therefore be considered as one of the most
preferred strategies in terrorism and insurgency as it increases the chances of success
without defeating the enemy’s forces.

3.4 Network-Centric Warfare

The British Royal Navy was the first force to embark on what today has become
known as “network-centric warfare”. The idea goes back to Admiral Sir John
Fisher—Mediterranean Fleet commander from 1899 to 1902—who imagined a
commander to be sitting right in the middle of a Clausewitz-like, i.e., rather “foggy”
picture. He coined the term “picture-center warfare”, thereby stressing the need to
be able to peep through that fog and thus be able to predict the enemy’s fleet’s
movements and thereby gain a tactical advantage.
The origin of his deliberations was strategic, though. While the Royal Navy
was more powerful than any of its potential enemies, it was inferior against any
combined force. Fisher, probably rightly, assumed that he could defeat either the
French fleet or the Russian fleet, but he would stand no chance if the two would team
up. The only way to avoid the possibility of being defeated would be to meet each
enemy’s fleet before it could join the other. In other words, he needed to predict his
20 3 Strategies and Tactics

enemies’ movements. Fisher reckoned the chances of winning to rely on overcoming

the asymmetry of information. Obviously, though, it would hold true for any kind of
tactical engagement as well. In order to tackle the lack of information, Fisher created
a new kind of naval intelligence service: it was to be the “all-seeing eye” which later
on, with the advent of wireless communication, would even enable the commander
to get more or less real time picture of events at sea. Taking in combination with the
increase of the speed of the British fleet the central problem of the British navy at
that time, namely their lack of resources to provide sufficient cover for all contested
areas, would be solved (Friedman 2009, pp. 3–10).
Efforts to obtain a better picture of the battlefield were undertaken by other forces
as well. One example from WWI was the use of balloons as observation post by
both the British and the German armies. The observer was located in a basket
suspended beneath and could, by means of a telephone, both identify targets for
artillery attacks at a greater range than it would ever have been possible from the
ground and subsequently deliver accurate reports on the bracketing of the enemy
forces, effectively transitioning artillery attacks from direct to indirect fire (Grattan
2009).
Today, air missions play a major role in almost every military operation.
And even if it is obvious that information superiority is essential to have aerial
supremacy, gathering it is only the first step to succeed in hitting targets on
the ground. While so-called reconnaissance missions aim to collect information
about where, when, how and why one could engage the enemy, the information is
converted into knowledge assisting command by intelligence missions. In addition,
other missions such as surveillance are dedicated to force protection and the
detection, identification and location of a target to allow airpower engagement.
Today, network-centric warfare (NCW) has taken over from picture-centric
warfare, taking into account the information-technology related changes within
the military. The Network Centric Warfare report to Congress in 2001 stated that
“The challenge for DoD4 is to harness the power of information technologies to
develop concepts of operation and command and control approaches that will be
information-driven rather than uncertainty-driven“ (Department of Defense 2001,
p. 2-2).
In their fundamental seminal work Network-Centric Warfare: Its Origins and
Future (Cebrowski and Garstka 1998) emphasize the capabilities of computer
networks to deliver integrated pictures of the battle and the importance of making
it available to all levels of command and control, including providing real-time
information on the positions of not only enemy troops by means of so-called red
force tracking, but of friendly troops through blue force tracking. Network-centric
operations can thus be understood as a kind of natural technology-based evolution
of Fisher’s all-seeing eye.
Following Cebrowski and Garstka’s publication, the Joint Vision 2020: Amer-
ica’s military preparing for tomorrow was published in the summer of 2000. It

4
DoD is the established abbreviation for “the United States Department of Defense”.
3.4 Network-Centric Warfare 21

was to lead the transformation the US Armed Forces that was needed to meet the
challenges of future warfare. In particular, US forces were concerned about the rise
of non-state actors endangering US forces in a more globalized world by having
easy access to ever cheaper information and weapon technologies (see, e.g., Joint
Chiefs of Staff 2000).
In the same way that the operational level emerged as a response to the military
requirements of the industrial age, NCW emerged to cope with the challenges
brought upon by the information age. Dahl (2002) analyses the expectations that
were raised by NCW in bringing about a fundamental change in the context of the
operational art.5 NCW is considered to have its roots in the operational art since it
aims to facilitate integration and synchronization within the commander’s decisions.
However, Dahl (2002) points out that many of the aspects traditionally ascribed
to the operational art such as synchronization or simply dividing an operation
into different phases would turn to be much less important on a network centric
battlefield.
Information superiority does not mean to have more information in the sense
of knowing more than the enemy, but rather to have that kind of information that
would actually give rise to an advantage. Clausewitz was well aware of this as he
mentioned priority of quality over quantity when collecting information. One of the
Clausewitzean trinity, “the interplay of chance and probability” highlights precisely
this, the importance of obtaining “good information”. Moreover, many authors point
out the non-linear character of war, which Clausewitz illustrated with the concept of
“fog of war”.
Clausewitz’s non-linear concept of war has nothing to do with the concept of
non-linear battlefields, as noted in the Joint Publication 3-0 (Joint Chiefs of Staff
2011), that assume conflicts are no longer fought along traditional front lines but
rather tend to become operations to be launched from bases. For Clausewitz, the
non-linear nature of war refers to the inability to predict results analytically, as it
is not possible to establish a system obeying rules like proportionality and addition
due to war continuously changing its character in ways that cannot be predicted
(Beyerchen 1992, p. 62).
Thus, in the military context to enjoy information superiority would not mean
to be able to predict settings, but rather that one could be able to achieve a state of
imbalance in one’s own favor by using information-related operations. This state of
imbalance, though, would be transitory as the information that brought about the
change of state would never be perfect (see Joint Chiefs of Staff 2000).

5
The concept operational art was introduced in the 1980’s, a couple of years after the introduction
of the operational level in the US Army doctrine. Historically, the concept operational art emerged,
as it happened with the operational level, from the need of a concept to manage military actions that
since the Napoleonic Wars became larger in time and space. The Operational Art is defined by the
DoD as: “the cognitive approach by commanders and staffs—supported by their skill, knowledge,
experience, creativity, and judgment—to develop strategies, campaigns, and operations to organize
and employ military forces by integrating ends, ways, and means” (Joint Chiefs of Staff 2015,
p. 176).
22 3 Strategies and Tactics

The quote included in the 2001 Report to the Congress: Network Centric
Warfare from then battalion commander Lieutenant Colonel “Ric” Riera is a good
illustration of the significance of the technological development for information
when having to take decisions on the tactical level:
“As a battalion commander, I need to see platoons. I need to see what platoons are doing.
I don’t need to see all the things on the battlefield, just the things that are important to me.
That makes decisions easier. I had to fight in one direction and then turn and fight in another.
Two things enabled me to do that: the soldiers with their level of training, and this command
and control system that allowed me to make better decisions. It’s not perfect, but it’s a lot
better than I’ve ever had. It’s powerful stuff” (Steele 2001, p. 36).

The advantage attributed to information superiority arises from the ability to

anticipate the course of events and, from a tactical point of view, to have an
operational picture that enables commanders to make better decisions. It seems
to go without saying that information superiority is the way to achieve decision
superiority, which is after all the ultimate end of NCW. The Army-Field manual
FM-3 defines nine principles of war: objective, offensive, mass, economy of force,
maneuver, unity of command, security, surprise and simplicity (see Department of
the Army 2008, pp. A-1–A-3). NCW is considered to enhance the ability to achieve
every one of them (Department of Defense 2001, p. 3-19).
If NCW aims to gain information for better decisions to be implemented so
that the adversary’s ability to react is limited, we are implicitly expecting that
armies would enjoy a second mover advantage. Then, decision superiority should
be sufficient to change the conception of a battle from being a simultaneous game
into one of it being a sequential one. As in a simultaneous game the players involved
cannot observe their rivals’ moves before making their own choices, neither would
that be possible in the battlefield, but the one enjoying information superiority could
not only anticipate his rival’s probable move but also make it conditional on his own
strategy.
A final aspect of NCW concerns the relationship between NCW and resource
allocation. One of the principles of war, the economy of force, states that a
commander should “use as little capacity as possible on aspects of the battle that
are not central to the objective” (Department of Defense 2001). The idea took a
leading role during Donalds Rumsfeld’s time in office as the United States Secretary
of Defense. Rumsfeld’s transformation of the US military—what soon was to
become known as the Rumsfeld Doctrine—consisted mainly in the adoption of
NCW, assuming that the use of information technology and airstrikes would reduce
the number of ground forces required in the theater of war. For this purpose he
created the Office of Force Transformation (OFT), appointing Cebrowsky—who was
considered the father of the NCW concept—as its first director.6
The most recent illustration of the Rumsfeld Doctrine is the role information
and airstrikes played in the Afghanistan War, when the operations to recapture the

6
The OFT was closed in October 2006, just 1 month before Donald Rumsfeld was followed by
Robert Gates in the Secretary of Defense.
3.4 Network-Centric Warfare 23

cities of Masar-e Sharif, Kabul and Herat from the Taliban relied heavily on the
accuracy and coordination of airstrikes. Information improved the accuracy and
coordination of airstrikes. It was mainly achieved by the role assigned to the Joint
Terminal Attack Controller (JTAC) who effectively became the commander of the
air forces involved. With forces on the ground relying heavily on air support, it
became necessary to add to the Forward Air Controllers (FAC)—who controlled
Close Air Support (CAS) from the air, primarily from light aircraft—its contribution
on the ground, the JTAC. Even though the attacks came from an air platform,
they were controlled from the ground, increasing precision and reducing the risk
of collateral damage.
Chapter 4
Conflicts as Zero-Sum Games

4.1 Asymmetric Conflicts

Over the past decade, the term “asymmetric” has begun to convey two meanings
within the military field. First, it has been used to refer to a special, i.e. “asymmet-
ric”, kind of threat that countries are facing, and second, it describes the military
conflicts resulting from these threats.1 It thereby refers to the nature of the conflict,
and even though the sources of asymmetry in a conflict are varied, usually it is the
asymmetry of force that is responsible for all others. On the other hand, though, it
must not be used in a too narrow sense: as in real life any two opposing forces most
likely command unequal defense budgets and/or are composed of different kinds of
units, military conflicts could of course be described as being “asymmetric”, except
that then examples for “symmetric” conflicts would at best be extremely rare indeed.
In the history of conflicts, power—and the imbalance of it—has proven to be
crucial to understand decisions in the battlefield. Weak forces are compelled to use
surprise and innovative tactics to challenge strong adversaries. Legendary examples
are provided by the narratives of epic ancient battles, the core of which are the very
imbalances of power between armies. For example, according to Herodotus, the
Greeks with a significantly inferior army were able to achieve a sound victory over
the Persians at the decisive battle of Plataea (479 BC), that ended the second Persian
invasion of Greece. Although Herodotus’ figures are highly contested, as he may
well have inflated the amount of Persian troops—according to Herodotus 350,000
Persian soldiers stood against 108,200 Greeks (Herodotus 1926)—in order to make
the battle appear more epic, his narration of the battle is a good illustration of how
the weak force’s search of innovative tactics can overcome its power-disadvantage.
In the battle of Plataea, it was the surprise factor used by the Greeks who, by faking

1
For an assessment of the use of the term “asymmetric” in military literature see Blank (2003).

© Springer International Publishing AG 2017 25

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_4
26 4 Conflicts as Zero-Sum Games

their own retreat, were able to confine the superior Persian army to its camp where
they could be fought on more favorable terms.
At least intuitively, though, whenever the distribution of power follows a pattern
where we can clearly distinguish between a strong actor and a weak one we tend to
expect for obvious reasons that the weak actor would have less chances of winning
against a stronger opponent.2 Nevertheless, history—and not just the legends from
ancient Greece—provides almost countless examples where it is the weak actor who
turned out to be victorious over a stronger opponent.
The following presentation borrows heavily from Arreguín-Toft (2001) who
analyses the period between 1800 and 1998 from the Correlates of War—a project
that collects different data sets with historical data, from territorial change or world
religion to civil wars and battles—finding that weak actors have won conflicts in
a far from negligible 29.2% of the cases and that the percentage of asymmetric
conflict victories has been growing steadily since 1800 up to reach 55% between
1950 and 1998. There are different arguments that may explain why weak actors win
asymmetric conflicts and why the amount of such victories is growing. Mack (1975),
for example, argues that there is a direct relationship between power asymmetry
and interest asymmetry. The greater the power disparity, the more vulnerable strong
actors are. Weak actors will have a higher interest in winning, since—in contrast
to strong actors—it is their very survival that is at stake. Arreguín-Toft (2001), on
the other hand, uses strategic interaction for predicting asymmetric outcomes and to
explain the trend of weak actors winning conflicts.
Arreguín-Toft’s model of strategic interaction describes ideal-type strategies for
each force in a conflict. Before going deeper into the description of the strategies,
some assumptions about the players’ types and their strategies are needed. First, we
assume that a conflict starts, once a military deployment takes place. Even though
we often observe that conflicts are seemingly triggered by an uprising or by an
indiscriminate attack against civilians like, e.g. the attacks on the New York World
Trade Center on September 11, 2001 (hereafter: 9/11), it is the strong force on whose
reaction it will depend whether the original “trigger” actually triggers a military
conflict involving the deployment of military forces on foreign soil or the projection
of military power into foreign territory, thereby only then leading to the outbreak of
an armed conflict. For that reason we will assume the strong force will always be
the “attacker”.3 Second, the strategies considered here will be rather war-winning
than war-terminating ones.4
On this basis, each force will have two possible options, respectively: the
“attacker” can either use a classical “direct attack” or it can turn to what is called

2
There are different approaches for power; the established practice uses a proxy for power where
a strong actor is defined by having ten times more material power than his adversaries. Material
power would be the product of state population and armed forces. See Arreguín-Toft (2001, p. 94)
and for empirical and quantifiable measures of relative power see Nutter (1994).
3
Of course there are cases where the weak actors starts the conflict, see Paul (1994).
4
Arreguín-Toft (2001, p. 100) finds that only in 4.1% of the cases the outcome was affected by a
conciliation strategy which include the use of bribes, offers of amnesty, power sharing etc.
4.1 Asymmetric Conflicts 27

“barbarism” while the “defender” would have to decide between a “direct defense”
and a “guerilla warfare strategy” (GWS). A short explanation of Arreguín Toft’s
four ideal strategies is as follows:
Direct attack Direct attack is the use of military force to win the war and destroy
the adversary’s capacity to resist. “Blitzkrieg” and “war of attrition” could thus
be considered examples for direct attack strategies.
Barbarism Barbarism is defined as the systematic violation of the laws of war
with a military or political objective. Because these strategies also look for
breaking the will of the adversary they would include actions that will rather
target non-combatants than military forces. Examples of barbarism are the use of
concentration camps or the strategic bombing of civilian targets.
Direct Defense Direct defense includes all military forces used to thwart an
attack from adversary forces.
Guerrilla Warfare Guerrilla warfare strategy arises when some significant part
of the society is organized into cells with the objective of imposing casualty costs
on the adversary, i.e. the invader, without any direct confrontation. The main goal
of guerrilla warfare is to break the will rather than the capacity of the attacker; the
latter would of course, due to the asymmetry of power, be pretty much impossible
anyway.
Arreguín-Toft (2001) argues that, because every strategy has an ideal counter-
strategy, if one is able to predict his adversary’s strategy, he would therefore improve
his chances of victory. He argues that, given the nature of these four strategies—the
direct ones aiming to destroy the capacity of fighting and the indirect ones aiming to
put an end to the will to fight—, in cases where both sides follow the same approach
it would most likely result in the weak actor’s defeat. This defeat is explained by the
striking power’s disadvantage with respect to the strong actor. By contrast, opposite
approaches would more likely result in the strong actor’s defeat as with opposite
approaches the power advantage is more likely to be deflected or dodged.
Arreguín-Toft (2001) tested his hypothesis using data sets from the Correlates of
War and found that in the conflicts from 1800 to 1998, strong actors won in 76%
of same-approach interactions while weak actors won in 63% of opposite-approach
interactions (Arreguín-Toft 2001, p. 111).
Using the same assumptions as Arreguín-Toft (2001), we can turn to game theory
in order to “confirm”5 the empirical results by an underlying theoretical model. The
strategic interaction model can be interpreted as a zero-sum game of two players
with two possible strategies. The resulting 2  2 payoff matrix will assume just two
feasible and mutually exclusive outcomes: either one wins the war, or one loses the
war. We will therefore assign to each player respective outcomes of 1 in case of
winning and -1 in case of losing.

5
As a famous quote, often attributed to Walter Heller, former chairman of the Council of Economic
Advisers, goes: “[Economists are] people who see something work in practice and wonder if it
would work in theory.”
28 4 Conflicts as Zero-Sum Games

Fig. 4.1 Payoff matrix for Weak

the strategic interaction
model Direct Indirect (GWS)
-1 1
Direct
1 -1
Strong
Indirect 1 -1
(Barbarism) -1 1

The corresponding payoff matrix of the strategic interaction model will thus be
given in Fig. 4.1.
In case both forces have to decide on their strategy without knowing about
their opponents’ choice, it would be a simultaneous game. No dominant strategies
would exist and neither would a Nash equilibrium in pure strategies, thus leaving a
uniquely determined Nash equilibrium in mixed strategies with each force playing
each of its strategies with the same probability and receiving the same expected
payoff equal to zero. At the same time, the model would predict that victories would
be found to be equally distributed across weak and strong forces. It would not be
consistent with the results observed from the “Correlates of War”.
However, as Arreguín-Toft (2001) assumes the strong actor to be the one who
initiates the conflict, i.e. the “attacker”, the assumption of a simultaneous game
looks less convincing. Instead, it suggests the strong force to be considered a first
mover, thus turning the game into a sequential one. Then, however, it becomes
immediately obvious that the second mover, i.e. the weak force, would always win,
i.e. the game would now generate a second-mover advantage. Therefore, if both
actors are at liberty to decide whether to follow a direct and an indirect strategy, the
force starting the conflict would always lose and the weak actor would always win.
Unfortunately, though, this result is not consistent with the empirical observations
either. The only way to “produce” Arreguín-Toft’s results would therefore be to have
another look at the “be-at-liberty” assumption.
Looking again at the original “trigger”, i.e. a weak actor initiating a process by
some act of terrorism or surprise annexation of territory at the end of which a full
blown military conflict may have evolved, quite often these original “triggers” have
arisen almost right out of the blue. Then, “attacking” does not necessary mean
that one is indeed the first to be able to pick a strategy one wants to play, and
neither may the “defender” be able to have the means to choose a best reaction
strategy. When a conflict breaks out—and this may happen quite sudden, too, even
though it had been “triggered” by the weak force: one only has to remember the
start of the invasion into Afghanistan in October 2001 which found the Taliban
rather unprepared and therefore immediately on the run—the weak force will not
necessarily have the time to put its best strategy into place, irrespective of whether it
is guerrilla warfare or direct defense. Therefore it may well be the stronger force that
effectively, i.e. in the game-theoretic sense, becomes the second mover and thereby
improving his chances of victory. Weak forces would only win in cases when the
strong force misreads the defender’s options. However, while it would explain the
4.2 Zero-Sum Games 29

percentages of asymmetric conflict victories in the first 250-year periods, the results
of the following periods, i.e. the trend of weak actors winning asymmetric conflicts,
still look puzzling. In particular, one would expect strong actors, due to advances
in intelligence and surveillance technologies, to become ever better informed over
time and their winning percentages to increase over time as well.
To explain the trend of weak actors winning more and more conflicts, Arreguín-
Toft (2001) argues that after WWII two socio-economic blocks emerged, showing
different patterns of socialization: the blitzkrieg pattern and the guerrilla warfare
pattern. It suggests that actors would, once they are on the threshold of an armed
conflict, not be at liberty to choose a strategy. First, resources, equipment and
training for a direct or an indirect approach are not interchangeable, and second,
actors might prioritize some kinds of threat over others and embark on supporting
strategies suitable to win the kind of conflict they expect. The increase of weak
actors’ victories could then be seen as an increase of interactions between these two
different patterns of socialization.
The laws of war have been evolving over time. Following the Hague Conventions
of 1899 and 1907 and the two additional protocols of the Geneva Conventions of
1977, the option for the strong player to play barbarism became constrained while
the weak player, not bound by the Geneva Convention, was still free to engage in
Guerrilla Warfare, leading to one of the two opposite-approaches interactions. It
would not only explain the higher percentage of weak actor victories in the last
of the period analysed, i.e. from 1950 onwards, where weak actors achieved 55%
of the victories in asymmetric conflicts, but would at the same time give a taste
of the future: attempts by weak actors to “trigger” military conflicts by initiating
mere incidents can be expected to be on the rise because they would more likely
than not be able to win the kind of subsequent military conflict and/or would “at
worst” achieve a propaganda victory if the strong actor decides not be tempted into
a military response.
Ultimately, though, strategic interaction alone will not explain the outcomes
resulting from asymmetric conflicts. It is rather the flexibility to choose a strategy
and the ability a force has to become second mover that better explains results in
asymmetric conflicts.

4.2 Zero-Sum Games

If we look back on military history, it might seem obvious that conflicts were
traditionally modelled as zero-sum-like games. Any gain achieved by one force
will account as a loss for the other one. There is a tendency to intuitively consider
information to play a crucial role when it comes to making decisions. Each force
will supposedly enjoy an advantage if it manages to anticipate the moves of its
opponent. In other words, it is expected that the outcome any force would receive
by knowing its opponent’s decisions in advance should be better than the one it
receives in absence of such kind of information and also better than the one when
30 4 Conflicts as Zero-Sum Games

Fig. 4.2 Payoff matrix for Red

zero-sum games R R
R R
B

Blue
R R
B

it is the other force which anticipates its moves. Otherwise, there would not be a
great incentive to gain information about any enemy’s intentions. It is the kind of
advantage suggested by Clausewitz’ famous “fog of war” image.
Technically, the zero-sum hypothesis may even be weakened by only assuming
that the ordering of one player’s payoffs is just the opposite of the other one’s
payoffs, i.e. whenever the first player prefers one outcome to another, his opponent
would prefer the latter to the former. Thus, rather than demanding payoffs to be zero-
sum, it would suffice that they are “quasi zero-sum”, i.e. every player’s payoffs can
be obtained by a strictly monotonous transformation of a set of zero-sum payoffs. In
order to keep the argument simple, the analysis is restricted to situations where each
force has just two strategies available, yielding to a 2  2 payoff matrix. Further,
none of the payoffs are equal, eliminating any kind of indifference. Following the
Cold War nomenclature, the forces are called “Blue” and “Red”, with respective
strategies B1 and B2 , and R1 and R2 . Denoting Blue’s possible payoffs by 1 , 2 , 3
and 4 , the “quasi-zero-sum” assumption implies that Red’s payoffs may be written
as R.1 /, R.2 /, R.3 / and R.4 /, with R being a strictly decreasing but otherwise
arbitrary transformation, i.e. R.i / < R.j / if and only if i > j . The resulting game
thus is given in Fig. 4.2.
Assuming 1 to be the best-possible payoff for Blue,6 there are only six possible
cases:
i) 1 > 2 > 3 > 4
ii) 1 > 2 > 4 > 3
iii) 1 > 3 > 2 > 4
iv) 1 > 3 > 4 > 2
v) 1 > 4 > 2 > 3
vi) 1 > 4 > 3 > 2
Cell-by-cell inspection easily reveals that in cases i) and iii) both players have
dominant strategies, while in cases ii) and iv) only one player has a dominant
strategy. All scenarios would result in a uniquely determined Nash equilibrium in
pure strategies. In cases v) and vi), only a mixed-strategy equilibrium exists. To

6
Otherwise, only the highest payoff for Blue can always be “moved” into the top left-hand cell by
swapping rows and/or columns.
4.2 Zero-Sum Games 31

discuss just one of each of the three possible outcomes, cases i), ii) and v) are looked
into in more detail (Figs. 4.3, 4.4, and 4.5).
Now, in order to test the “fog of war” hypothesis suggested at the beginning,
we turn to playing the game sequentially instead. In the four cases where at least
one player has a dominant strategy, the outcome would not change, i.e. the Nash
equilibrium in pure strategies is equal to the rollback solution, irrespective of who
starts the game. It is only in the final two cases, i.e. the mixed-strategy-equilibria
scenarios, that outcomes change: whoever moves second would receive a payoff that
is both better than the one he would get when moving first as well as the expected
payoff when playing simultaneously.
The result can be easily demonstrated with the indifference method. Taking
case v), if the game is played simultaneously, we will expect that Red will be
playing R1 with a probability q and R2 with a probability of .1  q/. Then, Blue
will receive an expected outcome smaller than the outcome resulting from moving
second, namely 4 :

EBBlue
2
D 3  q C 4  .1  q/ < 4 ; because 3 < 4 . (4.1)

Fig. 4.3 Payoff matrix for Red

case i) R R
R R
B

Blue
R R
B

Fig. 4.4 Payoff matrix for Red

case ii)
R R
R R
B

Blue
R R
B

Fig. 4.5 Payoff matrix for Red

case v)
R R
R R
B

Blue
R R
B
32 4 Conflicts as Zero-Sum Games

Likewise, Red’s expected payoff with Blue playing B1 with a probability p and B2
with a probability .1  p/, will be smaller than the payoff Red will obtain as second
mover in the sequential version of the game:

ERRed
1
D R.1 /  p C R.3 /  .1  p/ < R.3 / ; because R.1 / < R.3 / . (4.2)

Summarizing, what is striking, is that only in one third of all possible scenarios
it would be possible to enjoy an advantage by correctly anticipating the enemy’s
moves. In the majority of cases, collecting information about the enemy’s intentions
would not be translated into any better outcome than the one we have by not
knowing anything at all. Or, to put it in other words, information does not always
pay.

4.3 Max-Min Strategies and Mixed Strategies Equilibrium

with Ordinal Payoffs

So far, we have used the concept of a Nash equilibrium, and thus a best response
analysis. Alternatively, military commanders, because their decisions effectively
and greatly impact the lives of their subordinates, could be assumed to prepare for
the worst and try to minimize maximum losses, i.e. apply the so-called max-min-
method. Interestingly, though, the result would not change in the first four scenarios
discussed above, i.e. those zero-sum games exhibiting a unique Nash equilibrium in
pure strategies.
This is due to both players having so-called matched strategies: Had either Blue
or Red found out his opponent’s strategies—provided of course the opponent plays
the Nash-equilibrium-strategy—, he would not be able to achieve a better result. In
military manuals we often find the term saddle point to describe a two-player zero-
sum game equilibrium (see Wagner et al. 1999). The term refers to the graphical
depiction of any player’s possible payoffs for the different values of p and q, which
results in a saddle-shaped surface.
It is only in the two mixed-strategy scenarios where a switch to max-min-
strategies would lead to different outcomes. As no saddle point exists in these
scenarios, the result will no longer be symmetric, as can easily be seen from,
e.g. case v): provided both players go for the max-min approach, the equilibrium
strategies would be B1 for Blue and R2 for Red, thus resulting in the second-best
payoff for Red but only the third-best for Blue. In case vi) it would of course just be
the other way round.
Interestingly, though, neither player would fare worse than what he expected to
be the best of all his worst payoffs, if his opponent had known about him being
risk-averse, and thus enabling him to become a second-mover: To take case v), if
Red had anticipated Blue being risk averse, the result would not change; if Blue had
anticipated Red being risk averse, Blue would in turn respond by playing B2 and
4.3 Max-Min Strategies and Mixed Strategies Equilibrium with Ordinal Payoffs 33

achieve a higher payoff 4 > 2 but Red would receive no worse result from he had
prepared for, i.e. R.4 /.
If we compare the payoffs resulting from both players being risk averse with
those from just one of the players being risk averse, but his opponent not knowing
about this and thus sticking to the mixed-equilibrium strategy, we find similar
results: Due to it follows that and

1 > 2 and R.4 / < R.2 /;

EBBlue
1
D 1  q C 2  .1  q/ > 2 (4.3)
ERRed
2
D R.2 /  p C R.4 /  .1  p/ < R.2 / .

Blue would be better off by mixing his strategies, but Red would prefer facing a
risk-averse opponent.7
Summarizing, if at least one of the two opponents embarks on a loss-minimizing
strategy, the chances of a second-mover advantage to materialize become even
smaller.
So far, we implicitly had to assume cardinal payoffs, as otherwise no mixed-
strategy equilibria could be computed. However, if we think about battles, it is
easy to see that forecasting eventual outcomes linked to every strategy would be
quite unrealistic. Moreover, if we aim to use cardinal payoffs we need perfectly
defined objective functions capable to deliver results. These objective functions
should include information about casualties, costs, opportunity costs, etc. It would
require not only to have all relevant information but also to have a means by which
to numerically compare the different dimensions of the respective objectives, i.e. to
have a proper weighting available. Instead, one would have to be content with being
able to just rank the results expected to follow from every strategy combination.
Still, even though no cardinal values can be attached, the problem of not being
able to compute mixed-strategy equilibria can still be overcome by using a method
suggested by Haywood Jr. (1954) when analyzing the WWII “Avranches Gap”
battle. Drawing on the battle report by General Omar Bradley (Bradley 1951), “A
Soldiers Story”, to explain the battle that took place in Avranches in August 1944,
where General Omar Bradley commanded the US reserves in the fight against the
German Ninth Army, he first listed all outcomes General Bradley could possibly
expect depending on the American and German strategies by attaching ordinal
values.
While the Germans had two possible strategies, to attack .G1 / or to withdraw
.G2 /, the Allies had three options: to reinforce the gap and trap any attacking
German troops .A1 /, to move the reserve in order to disrupt the German withdrawal
.A2 /, or to keep the reserve waiting for one more day, hoping that more information
on the German strategy comes up .A3 /.

7
Similarly, we will find in case vi), Red would be better off mixing his strategies and Blue would
prefer a risk-averse Red.
34 4 Conflicts as Zero-Sum Games

The battle could thus have had one of six different outcomes. Haywood depicts
each of Bradley’s strategies by using vertical lines in order to rank the respective
results. Each line will then contain two points representing the two possible resulting
outcomes for each of the two German strategies. The lower value in each vertical
line will show the minimal outcome Bradley will achieve when playing that strategy.
The outcome resulting from Bradley mixing his strategies can then be shown by
the intersection points of the lines connecting the outcomes resulting from each
German strategy. The graphic can thus illustrate the expected outcomes when
Bradley uses the max-min approach in pure and in mixed strategies. As Bradley has
three strategies, Haywood, in order to avoid a three-dimensional graphical display—
i.e. to depict intermediate positions between A1 and A2 , between A2 and A3 , and
between A3 and A1 —has to depict one of Bradley’s strategies twice. The first and
the last vertical lines will both describe the same strategy and therefore lead to
the same possible outcomes. The lines connecting the resulting outcomes by the
Germans playing either G1 or G2 do not follow a quantitative pattern, as payoffs
are not assumed to be cardinal. Still, even though the purely ordinal ranking makes
it impossible to compute Nash equilibria in mixed strategies, Haywood’s graphical
analysis allows to compare the respective max-min results in pure and in mixed
strategies.
The resulting graphic will be as follows (Fig. 4.6).
If Bradley is playing max-min strategies, he will choose the vertical line with the
highest of all the lowest outcomes, as it would be the payoff he would be guaranteed
to achieve whatever the Germans opt for. One can thus see that Bradley’s max-min
strategy would be to play A3 , resulting in his third-best payoff, which is “Moderate
pressure on withdrawal”.

Possible pressure on withdrawal

Strong pressure on withdrawal

Estimated Outcomes

Moderate pressure on withdrawal

Weak pressure on withdrawal

Gap holds

Gap out

A A A A
Bradley’s Strategies

Fig. 4.6 Mixed strategies for the Avranches Gap Situation; adapted from Haywood Jr. 1954,
p. 379. The dashed line represents German strategy G1 , the solid line represents German
strategy G2
4.3 Max-Min Strategies and Mixed Strategies Equilibrium with Ordinal Payoffs 35

However, Bradley could still fare better by selecting a position between two of
his pure strategies. He would still rule out strategy A1 as both respective outcomes,
and thus any combination of them, would be worse than the max-min outcome in
pure strategies. Bradley could therefore mix his strategies by never playing A1 and
playing A2 and A3 with a probability of 12 each. The expected resulting outcome
would be located somewhere between the second and the third best results—because
of the ordinal ranking one could not know exactly where—, thus leaving Bradley
better off compared to playing any pure strategy.
In the cases considered in our model—cases v) and vi)—we have two players
with only two possible strategies each, making the graphic depiction much simpler
than that of Haywood. Again, each player can compare the resulting outcomes he
can expect from playing each of his pure strategies with the one he will receive
by mixing strategies along the vertical lines. The graphic will consist of two single
vertical lines, one for each strategy, and five possible outcomes: four resulting from
the two by two combinations of pure strategies and the one connected to playing
mixed strategies. This would be found in the point where the lines that connect the
outcomes resulting from each of the enemy’s strategy intersect.
In order to illustrate the argument, consider, e.g., case v). Blue would compare
his expected max-min outcomes of playing pure and mixed strategies, depicting
his strategies B1 and B2 by two vertical lines, each vertical line ranking the two
respective outcomes. The curves between the vertical axes will connect the possible
outcomes resulting from each of Reds strategies (Fig. 4.7).
For any particular vertical line, or pure strategy, we know that Blue would be
guaranteed the lower of the two values. The highest of the two lowest outcomes
Blue’s Outcomes

B B
Blue’s Strategies

Fig. 4.7 Mixed strategies for ordinal payoffs, case v. The dashed curve represents Red’s strategy
R1 , the solid curve represents Red’s strategy R2
36 4 Conflicts as Zero-Sum Games

would be achieved by Blue playing B1 , yielding 2 . However, we can easily see that
Blue can expect some outcome higher than 2 by mixing his strategies.8
We therefore find that the evaluation of ordinal payoffs leads to the same
conclusion as that of cardinal payoffs in the sense that Blue would be better off
by mixing his strategies, but that Red would prefer facing a risk-averse opponent.

8
We observe that the resulting expected outcome by mixing the strategies lies between the second
worst and the second best possible outcomes. This result is consistent with a “Folk Theorem” long
known to game theorists before it was actually published for the first time: In repeated games with
mutual cooperation with players announcing their strategies, each player will receive their Nash
outcome unless one player deviates in which case all players will play max-min, with the max-min
outcome being slightly worse than the Nash equilibrium.
Chapter 5
The Advantage of Moving First Versus
a First-Mover Advantage

In game theory, a player enjoys a first-mover advantage if he achieves a higher

payoff by turning the game into a sequential one with him being the first mover,
provided of course that the game can be changed in the first place. Leaving the
proviso aside, once we are talking about tactics such a statement might sound
obvious rather than revealing, though. Jensen (2015) argues that, if in a battle what
can be seen can be hit, military actors should be primed for the offense, i.e. trying to
make the first move by striking first. It holds true in particular in naval engagements
and aerial dogfights, and obviously it is the lore told by countless Wild West movies:
he who draws first wins.
However, striking first can be considered advantageous from a broader point of
view, too. Take for example so-called pre-emptive or preventive attacks. An attack is
called pre-emptive, if a country assumes that an enemy is about to attack: by striking
first, the enemy would—hopefully—deny the other party the ability to attack. A
preventive attack, on the other hand, is used to counter less immediate threats. Both
are sometimes considered to constitute self-defense actions (Mueller et al. 2006).
Although both pre-emptive and preventive attacks are based on the ability of
striking first, they can be interpreted as defense strategies, and as such the act of
defense per se suggests a second mover nature. In terms of offense vs. defense the
question is which strategy, i.e. striking first or upholding enemy assaults, results in
better outcomes. Jervis (1978) defines an offensive advantage as a situation where
destroying and holding the enemy’s territory is easier than defending one’s own.
On the other hand, a defensive advantage occurs if protecting and holding one’s
territory is easier than moving forward. Quite often both technology and geography,
i.e. terrain, are decisive factors for determining whether to find the advantage in
the offense or in the defense (Jervis 1978, p. 187). Furthermore, historians have
argued that it is the very nature of the kind of weapons that have been procured
and the military technology that has been invested in that will determine the overall
outcome. The more suited they are for offense purposes, the higher the likelihood

© Springer International Publishing AG 2017 37

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_5
38 5 The Advantage of Moving First Versus a First-Mover Advantage

of war will become, while if their primary use would be defense, the likelihood of
war should decrease (Levy 1984, p. 219).
One cannot completely rely on technology, though, to distinguish between
offensive and defensive measures, as often weapons can be used for both pur-
poses. Moreover, it is rather the objectives than the means used that differentiate
one from the other. Yet, Levy (1984) offers an interesting classification of the
offensive/defensive balance of military technology over the last eight centuries.
The supremacy of one over the other will depend largely on the developments
in the conduct of warfare throughout history: siege tactics, the success of the
longbow, the use of artillery or developments on the fortification’s construction
have determined how periods with an offensive predominance alternated with those
where the defense played a more important role. Levy’s analysis focuses on the
attempts military historians and other authors made to identify the offense/defense
balance of the 450 years preceding the end of WWII. He finds two periods adding up
to a total of 55 years where there has been a consensus on a superiority of offensive
strategies. On the other hand, the dominance of the defensive, also to be divided
into two time periods, accounted for 130 years. The remaining 265 years remain
unclassified because the diametric positions of the references taken into account
prevent a consensus on its classification (Levy 1984, pp. 230–234). This study finds
a historical trend in favor of the defensive. It is in line with many authors throughout
the history of military thought, such as, e.g., Clausewitz who considered defense to
be the form of warfare most likely to carry the victory.
However, being the first mover in a game-theoretic context is not equivalent to the
military concept of moving first. Aside from offense/defense considerations, moving
first in a military context is almost exclusively referred to as striking first, while in
game theory we find other ways that can lead to a first mover position: either one
actually anticipates the move of the other player or one manages to signal that there
is no possibility to deviate from the commitment of the strategy announced.
The biggest difference, though, is that commonly in the military literature any
advantage given by moving first is assumed to be an advantage that comes from
creating a disadvantage for one’s enemy. However, as we have seen in Chap. 4, in
quasi zero-sum games, i.e. the one kind of game where making oneself better off
automatically implies that one’s opponent will be worse off, there can never exist a
first-mover advantage. Thus, a game-theoretic first-mover advantage is only about
making the game proceed along a path different from the one preferred by one’s
opponent in order to make oneself better off, irrespective of whether this creates an
advantage or a disadvantage for the enemy.

5.1 Anti-Access/Area Denial (A2/AD)

A classical case where one has indeed to be quick to become a first mover,
thereby enjoying a first-mover advantage, can be found in the field of Anti-Access/
Area Denial, or A2/AD, the standard military acronym. Whenever a country tries
5.1 Anti-Access/Area Denial (A2/AD) 39

to prevent its opponent to project power—in particular if that opponent has to

be considered to be significantly superior in terms of both force and military
capability—the best way would be to first try to prevent the enemy from getting
access to the battle space (“Anti-Access”), and/or—if that has failed—to deny
largely unrestricted movement within that battle space (“Area Denial”). This holds
irrespective of whether the country employing an A2/AD strategy is the initiator of
the conflict by, e.g., attacking its neighbors and trying to prevent that any country’s
strong ally would rush to its help, or whether it finds itself on the receiving end of
an opponent’s aggression.
Technology has enabled countries to develop measures which, in the event of
a conflict, would hinder the deployment and movement of action of the enemy.
While the purpose behind any A2/AD strategy is far from being new—it is obvious
that every force in a conflict will do anything to deter the access and the ability
to maneuver of any adversary—technology has, over the years, developed in such
a way, that even states with modest defense budgets can obtain weapons that are
precise enough to be used from large distances, deterring the access of any enemy
and making A2/AD strategies a real challenge. The way wars are fought is directly
linked to the technology available, and although technological advantage contributes
to win wars, one can expect that sooner or later every technological advantage
would be countered. The invention of gunpowder, for example, meant one of the
most significant changes in warfare, but the capacity cannons had for destroying
fortifications was quickly followed by a change in the fortifications’ architecture,
which diminished their effectiveness (Keegan 1993, p. 321).
The sea has traditionally been regarded as the primary logistic means for military
forces, making the command of the sea the cornerstone of naval strategy. Therefore
any strategy aiming to deter A2/AD would have its roots in the art of naval strategy.
Command of the sea, or in technical terms sea control, is exercised in three ways:
by having the ability of attacking the enemy’s shipping and commerce, of striking
targets ashore and by launching amphibious operations to land ground forces in
the enemy’s territory.1 Although the theoretical concept of command of the sea
was developed when the sea was the only global common,2 today, air, space and
cyberspace have also become contested areas where naval strategies to sea control
are put in use.
The first part of A2/AD, i.e. anti-access, illustrated by the ongoing situations in
the China Seas, is to be discussed in more detail in Sect. 5.3. An example for the
second case is the prelude to “Operation Sealion”, Germany’s attempt to launch
an invasion of Britain in the second half of 1940, during the so-called “Battle
of Britain”: by successfully denying the German Luftwaffe to freely access the
air space over Britain, the Royal Air Force negated the one prerogative without

1
In short sea control is exercised using the so-called three Bs: blockade, bombardment and boots
on the ground (Armstrong 2015).
2
Global commons is used to refer to domains or areas that are not under the control of any state
but on which all rely.
40 5 The Advantage of Moving First Versus a First-Mover Advantage

which any attempt to invade would most likely have turned out disastrous for the
aggressor, resulting in “Operation Sealion” being cancelled for good (see, e.g.,
Tangredi 2013 and McKinstry 2014). Most likely, though, it will be a maritime
surrounding where an A2/AD situation should be expected to evolve in the twenty-
first century. Nevertheless, even though many technologies likely to be used will be
very modern, one of the oldest to be found in naval warfare can still be expected to
form part of it: we talk about mining.

5.2 Mining

The history of explosive mines goes back to the early fifteenth century, when
Florentine agents placed an explosive charge in a passage in the walls surrounding
the city of Pisa (Department of Doctrine and Training Development 1976, p. 133).
The origin of naval mines can even be found almost a century earlier in China, where
they were known as the “submarine dragon king” (Needham et al. 1986, pp. 203–
207 and Turnbull 2002, p. 35). They have been routinely used since at least the
American Civil War.
Assume now Red to be the country considering mining in order to prevent Blue
to enter some maritime space. Further, it is assumed that Blue does not have a
successful anti-mining strategy, i.e. there is no way to gain access, once mines have
been laid, without unacceptable damage to its naval forces. Note, though, that mines,
as they cannot differentiate between friend and foe, would, once laid, present a latent
danger to friendly shipping, too.
Each country has two possible strategies: Red would have to decide whether
to lay mines while Blue must decide, given the tensions between both countries,
whether to try to secure access or refrain from doing so, leaving its ally to its
fate. The terms ACCESS/REFRAIN and MINING/NO MINING will describe the
respective strategies of Blue and Red. Then the payoff matrix will look as follows
(Fig. 5.1).
Red would find that NO MINING would dominate over MINING. If Blue plays
ACCESS, it implies that the mining strategy has not succeeded in deterring Blue
forces but would instead just led to higher costs for both countries. On the other
hand, if Blue plays REFRAIN, laying mines would not have been necessary but

Fig. 5.1 Payoff matrix for Red

the mining model
MINING NO MINING
R R
ACCESS
B B
Blue
R R
REFRAIN
B B
5.2 Mining 41

would have caused increased costs for Red, since it would have to remove the mines
in order not to hinder its own shipping.
Red’s payoffs can thus be ranked as

R4 > R3 > R2 > R1 . (5.1)

Blue would naturally prefer to have a mine free access. The highest possible
outcome would thus be found by playing ACCESS without facing any mines that
would restrict their movement of action. In case Red has placed mines, Blue would
be better off by playing REFRAIN. Summarizing, its payoffs are as follows:

B2 > B4 > B3 > B1 . (5.2)

The payoff distributions reveal that we are no longer dealing with a zero sum game,
then as B4 > B3 and R4 > R3 , a gain achieved by one player would not necessary
result in a loss for the other one. A cell-by-cell inspection reveals that there exists a
single Nash equilibrium in (ACCESS/NO MINING).
However, if we consider the nature of the two “aggressive” strategies available
to each player, MINING and ACCESS respectively, it is obvious that they are
extremely unlikely to happen at the same point in time, if merely for the fact that
mining simply takes time. Red has to decide in advance whether to place mines or
not, as otherwise, once Blue has managed to put its forces into position, mining
would no longer work as deterrent. However, if Blue decides to play REFRAIN,
there would not be a threat and Red would prefer to enjoy unlimited movements
for its own naval forces and civilian shipping alike. Therefore MINING has to be
played first, or not at all, as would ACCESS.
We therefore compare the two respective sequential games. If Red moves first
(see Fig. 5.2) we find mining would work, inducing Blue to refrain and making
the resulting outcome from playing MINING higher than the one when playing the
dominant strategy.
On the other hand, if Blue moves first, the outcome would be the same as in the
simultaneous game, implying that Red, if compared to the outcome when moving

Fig. 5.2 Game tree for the

mining model
42 5 The Advantage of Moving First Versus a First-Mover Advantage

first, would be worse off resulting in both a first-mover as well as a first-striker

advantage because Red simply has to hurry in order to have its mines in place before
Blue can contemplate whether or not to try to gain access.
A final point: It goes without saying that in war all payoffs would usually have
to include random variables. Therefore, the ranking of both the Bi as well as the Ri
refer to expected values and would not guarantee that some payoff has always to be
larger than some other payoff. Interestingly, though, while the US employed mining
to prevent German submarines from entering areas like the Chesapeake Bay on the
East Coast and the Florida keys, the policy immediately drew criticism once it took
its toll on US merchantmen—and even one US destroyer—even though in all cases
this was due to simple navigational errors. It led, however, to the abandonment to
a minefield off Cape Hatteras on the North Carolina Outer Banks only for the area
to become known as Torpedo Alley with more than 400 ships having been lost to
German U-boat attacks due to the lack of any effective barrier against such attacks
(Office of Naval History 1946, p. 116f).
Interestingly, the A2/AD scenario is equivalent to a strategy well known in
industrial organization, namely “Raising Rivals’ Costs”, or RRC in short. It refers
to strategies whose main purpose is to obtain a benefit from causing disadvantages
to rivals in—both potentially as well as actual—competitive markets.
The idea goes back to Director and Levi (1956), who suggested that firms with
monopoly power could impose coercive restrictions on suppliers and customers.
Later, Williamson (1968) provided an example with the case of a coal company
accused of conspiring with its workers’ union to raise the wage rate, showing how
an industry-wide wage contract could affect labor-intensive firms more severely and
force their exit while benefiting the capital-intensive competitor.3
In its most extreme form, RRC, though—and this the most illuminating example
used in respective textbooks in economics—, would increase both the incumbent’s
and any potential competitor’s cost in such a way that the competitor’s payoff
from competing in the market would become negative and would thus prevent the
competitor from entering the market in the first place, thus leaving the incumbent as
the sole supplier. However, it comes at a cost, as the incumbent’s profit, while still
being higher than a duopoly profit, would be reduced.4
The strategy would thus be dominated in the simultaneous game just the way NO
MINING was dominated. In order for RRC to be successful, i.e. generate a first-
mover advantage, it has to be played before any competitor has himself committed
to market entry, because otherwise it would fail to maintain one’s monopoly status
and would only make everyone worse off.

3
Salop and Scheffman also outlined the conceptual framework of raising rival’s costs to force
market exit. See Salop and Scheffman (1983) and Salop and Scheffman (1987). Further literature
on entry barriers based on increasing the costs of competitors can be found in Dixit (1979), and
Rogerson (1984).
4
It is for this very reason that the strategy is called “raising rivals’ cost” rather than “raising rival’s
cost”.
5.3 A2/AD and the China Seas: A Wider Perspective 43

5.3 A2/AD and the China Seas: A Wider Perspective

Today, the most striking A2/AD example is given by China.5 On the one hand,
China’s naval modernization signals its desire to develop a military force capable
of deterring US naval and air forces in case a conflict about Taiwan’s sovereignty
breaks out, and, on the other, the willingness to displace the US presence in the
Pacific.6 China has increased its military budget for the fifth consecutive year.7 Even
though, this should not be interpreted as exclusively related to the increasing tension
about Taiwan, but also as a consequence of China’s economic development and the
growing need of building a military force/presence responding to the expectations
of any world power,8 China’s increasing military tension with the US over the last
years can nevertheless not be ignored. Swaine (2015b) argues that this is a question
of power dominance over the Western Pacific. After World War II, the US policy
was based on both its leadership role as well as its dominant power, which in turn
persuaded many (not just Pacific) nations to enter a military alliance with the US,
which in turn formed the basis for a long-lasting period of peace and subsequent
economic growth in many regions of the world. From the 1950s to the 1970s China
was happy with the American dominance because it helped to offset the power of
the Soviet Union and at the same time facilitated economic growth in its Asian
neighbors, avoiding arms races and, at least for the time being, historical rivalries.
However, China’s economic success, the collapse of the Soviet Union and China’s
fear that its regime could be the next to fall, ended the Chinese toleration of the US
dominance of the Western Pacific (Swaine 2015a). The Chinese territorial claims
in the South China Sea recently peaked with China’s island-building project: while
China claims that the islands are situated within its territory, its claims are based not
only on archaeological finds but on the fact that no one had contested a document
released by China in 1947 showing the infamous Nine-Dashed Line. This document
has since not only been contested by a number of countries from the Philippines
to Vietnam,9 but also contradicts the Convention of the High Seas as agreed on the

5
China itself does not use the term A2/AD but counter-intervention. See Fravel and Twomey
(2015).
6
An evidence that the A2/AD strategy from China not only includes military objectives related
to Taiwan would be China’s acquisition of aircraft carriers, which would not be needed in any
Taiwan-only conflicts because Taiwan lies well within land-based Chinese aircraft. See Lovelace
(2014, p. 167–168).
7
For 2015 it is expected that the military Budget will reach around $141,45 billion, 10,1% more
than in 2014 (Gady 2015).
8
In 2010 China’s then foreign minister stated “China is a big country, and other countries are small
countries, and that is just a fact.” Cited from Krepinevich (2015).
9
After a year of building artificial islands, China has recently announced the termination of the
project. The end of this island building project can also be seen as an attempt to alleviate the
tensions between the two countries on the verge of the US-China Strategic and Economic Dialogue
(S&ED), held in Washington in June 2015, and the first state visit of president Xi Jinping of China
to the US in September 2015.
44 5 The Advantage of Moving First Versus a First-Mover Advantage

1958 United Nations Conference on the Law of the Sea, or UNCLOS I, which was
ratified by, at the time of writing, 63 countries including China.
China’s A2/AD strategies would consist of the procurement of the technology
necessary to deter the access and freedom of movement in the China Seas. They
include a number of measures, ranging from air bases and naval facilities to counter-
maritime and counter-air systems, severely limiting US military freedom of action
within the Pacific area. In order to be able to cope with these challenges, in July
2009 US Secretary of Defense Robert M. Gates introduced the operational concept
of the so-called “Air Sea Battle”, or ASB. It aimed to employ means across all of
the five domains—air, land, sea, space and cyberspace—to overcome the access and
deployment obstacles imposed by A2/AD strategies. It was controversial, though,
right from the start, as critics claimed it to be nothing more than a component of
the DoD strategic mission to A2/AD threats,10 and in May 2013 even the ASB
Office effectively acknowledged this critique by releasing an unclassified document
defining ASB as “A limited objective concept that describes what is necessary for the
joint force to sufficiently shape A2/AD environments to enable concurrent or follow-
on power projection operations“ (Air-Sea Battle Office 2013). The original concept
of ASB did not last long anyway, as the development of ASB in the past few years
highlighted, among other things, the deficiencies of effectively relying on the Air
Force and the Navy to counteract A2/AD. On January 2015 the Pentagon abandoned
the name Air Sea Battle to introduce a new concept called “Joint Concept for Access
and Maneuver in the Global Commons”, or JAM-GC, which includes not only the
Air Force and the Navy, but also land forces.11 This new concept was presented as
the development of the ASB and includes all new operational requirements that have
emerged in the past 6 years. It can then be considered the command of the sea in the
modern world (Armstrong 2015, p. 16).
Not surprisingly, Beijing has since expressed concerns over these countermea-
sures several times and described both ASB and JAM-GC as a provocation against
the Chinese government (see, e.g., McCarthy 2010). One should note, however, that
while the US constitutes the only adversary China has in mind when designing
its A2/AD strategies, the US forces have to deal simultaneously with several
adversaries around the world such as, e.g., Iran building A2/AD capabilities for
future conflicts over control of the Strait of Hormuz.
Thus, as long as A2/AD strategies do not constitute an immediate threat, the
implementation of countermeasures may depend on the distribution of priorities
the force has among armed conflicts facing at a time. Following the Red and Blue

10
Some observers reported this new operational concept as a budgeting strategy to justify the
investment in weapons programs for the Air force and the Navy in view of the Congressional
hearings (Hoffman 2015).
11
For a wider explanation on the ASB concept development see Morris et al. (2015). Critics say,
though, that JAM-GC is nothing but the old ASB, with merely some token land elements having
been added in order to get the Army and the Marine Corps on board and avoid creating a picture of
the different forces being at war with each other when it comes to persuading Congress to release
the necessary funds (LaGrone 2015).
5.3 A2/AD and the China Seas: A Wider Perspective 45

Fig. 5.3 Payoff matrix for

A2/AD

nomenclature used before, and assuming that each force has two strategies available,
which are to either build A2/AD and, respectively, counter A2/AD capabilities, or
JAM-GC, or to maintain the status quo and remain passive, the resulting payoff
matrix would be given in Fig. 5.3.
We assume the ranking of the possible outcomes in such a scenario to be as
follows: provided JAM-GC is affordable, Blue would prefer any situation where he
can spare the threat of dealing with an A2/AD strategy, whether or not a military
confrontation—and thus the need for access and maneuver capabilities—looms. In
absence of an A2/AD challenge Blue would also be better off by the saving on JAM-
GC expenditures. However, if A2/AD arises, Blue would prefer to count on having
JAM-GC components in order to maintain its ability to project power. In short:

B4 > B2 > B1 > B3 . (5.3)

The best possible outcome for Red would be to enjoy the advantage of having
A2/AD capabilities and no effective counter strategy having been put in place,
followed by the status quo. In case Blue has adopted JAM-GC, the least Red would
want is to allow riskless operational freedom for Blue. The resulting payoffs would
thus be ranked

R3 > R4 > R1 > R2 . (5.4)

Again, i.e. just like mining—it would certainly form a part of A2/AD anyway—
the game is a non-zero sum game, due to both B4 > B2 and R4 > R2 as a gain
achieved by one player would no longer result in a loss for the other one.
Cell-by-cell inspection reveals that only Red has a dominant strategy—which is
to build A2/AD12 —and the existence of a Nash equilibrium with Red employing
A2/AD and Blue going for JAM-GC to counter the A2/AD threats. Even if we
assume Red and Blue being conservative in the sense of minimizing ones respective
possible losses by using max-min strategies, we find “matched strategies” giving
rise to the same solution as in the Nash equilibrium.

12
Unlike in the case where China’s A2/AD strategies consisted in mining, thus mining was
dominated by no mining.
46 5 The Advantage of Moving First Versus a First-Mover Advantage

Fig. 5.4 Game tree for

A2/AD

On the other hand, if we consider that any JAM-GC has to be interpreted as a

reaction to an A2/AD threat, the game would have to be played sequentially with
Red moving first. The corresponding rollback equilibrium is given in Fig. 5.4 result-
ing in an improvement for both players: Both players would maintain the status quo
yielding payoffs of B4 and R4 rather than B1 and R1 . In terms of game strategy, while
the dominant strategy would not be played by the first mover—as was the case with
the mining example—A2/AD would not be employed by Red, unlike mere mining.
Earlier, we have mentioned how improvements in technology enabled even
nations with comparatively small defense budgets to acquire A2/AD capabilities.
However, there are other factors contributing to the recent rise of such strategies. In
particular, Chinese A2/AD strategies emerged at the same time US military forces
were focused on both Iraq and Afghanistan.
In general, once the “strong” player has to deal simultaneously with the threat
of having multiple adversaries all over the world, the weaker one might well be
tempted to exploit the respective budgetary pressures and operational bottlenecks.
The US would then have to take into account the likelihood of future conflicts and
allocate respective funding, giving rise to further opportunity costs. In other words,
as the resources are limited, at least in the short run, the US might need to allocate
across both parallel and potential future conflicts.
To keep the analysis simple, we just assume two states of nature: either other
threats are conceived by the US as sufficiently small and would not be considered
a distraction when it comes to react to Chinese A2/AD policies, or other threats
require a reallocation of military assets because they are needed more urgently
elsewhere. The two cases would be distinguished by introducing a variable   0
to represent any eventual further opportunity cost of implementing JAM-GC to give
(Fig. 5.5).
As long as  < B1  B3 , the game is virtually the one discussed before: there
exists a uniquely determined Nash equilibrium given by A2/AD and JAM-GC,
respectively. If played sequentially, Red would begin by playing PASSIVE, with
Blue reacting by choosing PASSIVE, too.
In order for other conflicts to have any impact, these opportunity costs have to be
assumed to be sufficiently high to affect the ranking of B1 and B3 , i.e.  > B1  B3 .
5.3 A2/AD and the China Seas: A Wider Perspective 47

Fig. 5.5 Payoff matrix for

A2/AD with opportunity
costs

Now PASSIVE becomes a dominant strategy for Blue, yielding in turn a new
equilibrium for both the simultaneous and the sequential version with Blue relin-
quishing JAM-GC and Red implementing A2/AD. Note, that irrespective of the
state of nature, i.e. the magnitude of , Blue would always play PASSIVE in the
sequential game as long as Red knows about that state of nature. If not—Red could
be incapable of collecting enough information about the magnitude of , i.e. Blue’s
ability to deal with conflicts and its perception of the respective threats—, the best
Red can aspire to is to form an estimate on the probability p for  being small,
i.e.  < B1  B3 , in order to compare the expected payoffs playing A2/AD and
PASSIVE, i.e. if E.A2/AD/ D p  R1 C .1  p/  R3 and E.PASSIVE/ D R4 . If

R3  R4
p< , (5.5)
R3  R1

Red would choose A2/AD, otherwise it would remain PASSIVE.13

Then, as Blue would always prefer Red to play PASSIVE, Blue should commu-
nicate the truth if  is small, and it should try to coax Red into believing that  is
small even if it is not. Thus, the situation is equivalent to a classical signalling game.
For  being small the signalling costs should be close to or even exactly equal to
zero—all Blue has to do is to invite Red’s military as observers to some respective
exercise—implying it would always send a signal. If  is large, the result would
depend on the signalling costs of pretending to be able to fight two or more wars
at the same time without one operation compromising the other: If the signalling
costs are sufficiently large a separating equilibrium would ensue: It would not make
sense for a “weak” Blue to pretend otherwise, and Red would remain PASSIVE if
facing a “strong” Blue and employ A2/AD if facing a “weak” Blue. If a “weak”
Blue’s signalling costs are small, the result depends on p: it would either, in the
case of p being large, be a pooling equilibrium with the “weak” Blue successfully
masquerading as a strong one, or, in the case of p being small, it would result in
a semi-separating equilibrium, i.e. sometimes a “weak” Blue would pretend, and
sometimes Red would go for A2/AD even if a signal indicating a “strong” Blue had
been received.

13
Otherwise, the game could not be solved.
48 5 The Advantage of Moving First Versus a First-Mover Advantage

Time plays an important role in the accuracy of Red’s predictions: the more
time goes by, the better the information Red has for its estimations. And, in case
 is large, the more information Red can acquire and the more difficult it would
become for Blue to pretend otherwise. Red could thus choose to simply wait until it
is possible to make better estimations for , thereby getting a better understanding
of how likely it can expect Blue to invest in JAM-GC capabilities once Red has
started to build A2/AD capabilities.
The one obvious advantage of not waiting would be to eliminate the risk
of making wrong predictions. On the other hand, accepting the risk ensues the
advantage of being able to use the first move for negotiation issues. Consider, for
example, that Red gambles on the risk of underestimating Blue’s capabilities and
begins with setting up A2/AD facilities. At best, its estimations are right and Red
could thereby get a head start regarding its underlying expansion policies. At worst
its estimations are wrong and Red would have to deal with JAM-GC strategies. As
far as all these strategies do not imply the outbreak of a war or an insurrection,
JAM-GC strategies would only mean that Red’s implementation of A2/AD would
turn less useful than originally planned. It would result in a loss of investment, but
these sunk costs would otherwise have no impact on further decisions by Red. On
top of that, a similar if not much more expensive sunk cost would have been inflicted
on Blue raising Blue’s future opportunity costs as well.
Furthermore, already having an extra asset—i.e. A2/AD—in place before any
negotiations about underlying expansionist policies are to begin, could well change
negotiation terms in favour of Red.
In the China Seas, the development of China’s A2/AD capabilities could thus
be explained as both a result of China’s estimate of US goals and capabilities—in
particular the assumption that the US has got its hands full with maintaining its
superpower status—as well as against the background of China’s medium to long-
term strategies. With respect to the former, note that p may even not only refer to US
military capabilities, but to the willingness of the US to actually use it once China
actually begins to expand and threaten or annex territories because JAM-GC might
require pre-emptive attacks on Chinese mainland missile sites in order to eliminate
the threat of any crippling attack against US aircraft carriers trying to enter the
battle space. Then, the development of A2/AD capabilities might be interpreted as
China having been tempted, rightly or wrongly, to believe that the US would lack
the political will to enter into a war over sovereignty issues in either of the Chinese
seas, and/or that US foreign policy has simply failed to send a clear message that
there would indeed be a red line that must not be crossed.
Chapter 6
Irregular Warfare

War understood as two armies fighting against each other has become an obsolete
concept. During the last few decades we have been observing conflicts of a different
nature, not only because the actors involved are no longer legitimized by states, but
also because of the unconventional strategies utilized in particular on the side of
the non-state actor. Since 1975, the Uppsala Conflict Data Program (UCDP) has
been recording data from global armed conflicts confirming the increasing trend of
non-state conflicts.1
Kaldor (2007) uses the term “new wars” and “old wars” to distinguish irregular
conflicts from traditional ones. In contrast to conventional wars, where forces faced
each other over territorial claims, “new wars” have emerged in the context of
globalization.2 The decrease in inter-state conflicts is explained in Kaldor (2007)
by the increase in military alliances among states and arm control agreements
that have undermined the capacity to wage war against each other. At the same
time, the emergence of comparatively simple military technology such as IEDs has
enabled non-state actors to employ ever more kinetic means, thereby increasing the
asymmetry between armies.3
The new challenges of unconventional wars have been acknowledged by the mi-
litary mindset. British General Rupert Smith, Deputy Supreme Allied Commander
Europe (SACEUR) from 1998 to 2001, argues that interstate warfare no longer
exists. During his career in the British Army, the conflicts he faced were significantly
different from the ones for which military force was originally configured (Smith
2007). Taking for example the tank as the ultimate machine designed for full-scale

1
The UCDP defines non-state conflicts as the use of armed force between two organized armed
groups where neither of which represents a legitimate government of a state and which results in
at least 25 battle-related deaths in a year (Department of Peace and Conflict Research 2015).
2
On the topic of new wars see also Munkler (2005).
3
Hoffman (2007) uses the term “hybrid wars” to capture the property of intermingling attributes
these wars possess.

© Springer International Publishing AG 2017 49

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_6
50 6 Irregular Warfare

battles, we find just two significant tank battles4 over the last four decades: while
the penultimate one took place in the Valley of Tears during the Arab-Israeli war on
the Golan Heights, the final one—it was, though, the largest tank battle involving
US forces since World War II—was the battle of 73 Easting in the Gulf War in 1991
(MacGregor 2009).
Irregular warfare (IW), counterinsurgency (COIN), guerrilla warfare and asym-
metric warfare are terms often used interchangeably for describing conflicts dif-
ferent from traditional inter-state warfare. However, irregular warfare encompasses
all the others, i.e. the latter should be considered as mere aspects over the former.
Defining irregular warfare has been and still is confusing: some authors define it in
contraposition with regular warfare—overlooking the fact that there is no definition
of a regular war in the first place—while others focus on the objectives and operating
environment that characterize it. Within the US Department of Defense (DoD) two
alternative definitions were competing for official status until a final definition was
approved on October 2006 and subsequently used in most of DoD publications: “A
violent struggle among state and non-state actors for legitimacy and influence over
the relevant populations. IW favors indirect and asymmetric approaches, though it
may employ the full range of military and other capabilities, in order to erode an
adversary’s power, influence, and will” (Department of Defense 2007, p. 6; see also
Larson et al. 2008).
Acts of violence from non-state actors, terrorism and insurgencies have become
the main national security problem democratic states face, and an important area
military forces need to be prepared for.

6.1 Aspects of Irregular Warfare

According to the DoD definition of irregular warfare, population relevance, fighting

for legitimacy, asymmetric approaches and non-state actors cover the main aspects
of these conflicts.
Terrorism can well be considered as one of the “indirect/asymmetric approaches”
referred to in the above definition: it is indirect as it targets civil population instead
of armed forces, and asymmetric, as it will not be used by COIN forces. While some
authors like Kiras (2007) consider both terrorism and insurgency to fall within the
definition of irregular warfare as forms of violence where the acts committed are
legitimized by their political nature, only insurgency attempts to bring about change
through force of arms, and unlike terrorism it is supported by a significant proportion
of the population, though not necessarily for the right reasons: sometimes civilians
are simply coerced into “supporting” the insurgency out of fear of reprisals, which
has been acknowledged by authors like Butler and Gates (2010), who suggest a
more dynamic approach where terrorism is considered just “a strategy designed to

4
Tank battles as understood as two armies maneuvering against each other.
6.1 Aspects of Irregular Warfare 51

further a political agenda by a system of violence perpetrated by a non-state actor

against noncombatant targets thereby instilling fear and intimidation among a wider
audience” (Butler and Gates 2010, p. 12). Either way, terrorism would be seen as a
means to put pressure on governments. The view of terrorism as a coercive measure
implies that governments whose citizens are more vulnerable would more likely
be the subject of it (Arreguín-Toft 2001, p. 103). Terrorism as a strategy will be
confined to creating a psychology of fear by targeting civil population. On top of
that, when employing terrorism, insurgents can also take advantage of the media
unwillingly becoming, at least in those countries enjoying the freedom of the press,
supporters of the insurgency by both contributing to the atmosphere of fear and the
promulgation of the cause.
Another relevant aspect in irregular warfare is the use of violence against the
civilian population. Civilians are subject to violence on both sides of the conflict.
First—as noted before in the definition of terrorism—while civilians are the usual
targets in terrorist actions, it is the population of the so-called host nation in
particular that is likely to suffer violence from insurgent groups using it as coercive
means to increase support. Second, because COIN strategies, as will be discussed
later in more detail, which rely heavily on air operations and aerial attacks are
often unable to distinguish between insurgents and civilians, it is again the civilian
population that would be made to suffer.5 The most recent example of this can be
found in the Syrian civil war, where Amnesty International reported that aerial
attacks were indiscriminately used in some regions causing hundreds of civil
casualties (BBC news 2012). Finally, according to Butler and Gates (2010), irregular
warfare often shows episodes of opportunistic violence like private revenge or
robbery that are only possible because of the lack of constraints inherent to insurgent
organizations (Butler and Gates 2010, p. 7).
At the same time, violence against civilians inferred by the superior force plays
a very important role for insurgents’ recruitment. Due to information asymmetry,
insurgents are able to provoke COIN forces to engage indiscriminate attacks,
thereby either killing civilians or making the killing of civilians by insurgents look
as if they were caused by COIN forces. Then, victims’ relatives or friends might
be encouraged to seek revenge by joining the insurgency (see, e.g., Butler and
Gates 2010, p. 9). Insurgents, on the other hand, would be at liberty to either go
for individual targets, thus minimizing the civil casualties they cause in the host
nation, or to launch indiscriminate attacks.
The role of the population in insurgencies is not exactly new to modern conflicts.
Mao Tse-Tung illustrated the relationship between guerrilla and people as the
“fish” that swim in the “sea of popular support” (Tse-Tung 1937, p. 91–92).
For Clausewitz, public opinion was one of the centers of gravity in any popular

5
However, Butler and Gates (2010) noted that, once aerial bombardment became a means to inflict
economic damage on an enemy and/or to provide tactical support to ground troops fighting in
populated areas, the civilian population was bound to suffer even when the air force took great care
of not inflicting undue damage, implying that violence against the civil population has not been
confined to irregular warfare (Butler and Gates 2010, pp. 6–7).
52 6 Irregular Warfare

uprising (von Clausewitz 1976, p. 596). Support by the population has always
been a fundamental pillar of any successful insurgency as it not only guarantees
a constant flow of new insurgents but it becomes crucial when it comes to give
shelter to the insurgency. However, rather than relying on domestic support only,
insurgents would also look for external sympathizers and supporters. Currently, the
most striking example is the rising number of US and European citizens traveling to
join terrorist groups in Syria and Iraq. The US intelligence estimates that since 2011
more than 20,000 foreign fighters—with 3,500 coming from Western countries—
have travelled to Syria to fight for ISIS (Archick et al. 2015).
Thus, support by the population will not necessarily be obtained by coercive
means, as mentioned before, but also by legitimizing the use of violence. This is
achieved by displaying a moral superiority, for example by means of religion or
by exposing the current government’s corruption.6 At the same time, differences
in religion or ethnicity are often the source for inequality, which in turn has a
great mobilization power. Whenever one group feels discriminated by another
group, it might easily consider the use of violence as legitimate means. These
systematic inequalities between ethnic groups are often referred to as “horizontal
inequalities”. They describe inequalities between groups in contrast with the
“vertical inequalities” which refer to inequalities between individuals such as, e.g.,
income inequalities. The difference is important: while there is a positive relation
between the “horizontal inequalities” and the outbreak of a conflict, individual
inequalities are not found significantly related to the risk of violent conflict.7 Again,
one only has to look to the conflicts in Syria and Iraq, where religion and ethnicity
are responsible of the recent profound turmoil of the region.
Finally, there are other types of asymmetries characteristic of irregular warfare
giving an advantage to one or the other side of the conflict. The most evident
one favoring the COIN forces would be the asymmetry of conventional military
power and technology. On the other hand, the insurgents would benefit from
unconventional devices, not being bound by the Geneva Convention, and from local
geography, both in terms of territory and human geography.
In a 1997 speech to the National Press Club, General Charles Krulak, 31st
Commandant of the Marine Corps, talked about the reasons why the barbarians
overwhelmed the Roman Empire, in particular on how it was possible for a rather
backwater people to defeat the most civilized and technologically advanced power
of the time. Krulak (2000) explained in his speech titled “Not like Yesterday” that
the main reason was the ability of the Germanic tribes to neutralize the advanced
technology of the Romans “by luring the Roman heavy cavalry into marshes
where their mobility and shock power were rendered useless. They lead the Roman
bowmen into the forest where the trees negated the effectiveness of their arrows”.

6
Harmon (2010) finds that although terrorism has an undeniable political character, since the late
twenty century it has become more religious.
7
For horizontal versus vertical inequality, see Stewart (2000) and for statistical evidence of
horizontal inequality and violent conflict, Østby (2008).
6.1 Aspects of Irregular Warfare 53

Krulak (2000) pointed out that superior technology alone can not necessarily assure
victory for the “superior” force in irregular conflicts: Just as Roman technology and
doctrine was unable to cope with the advantage the Germanic tribes enjoyed when
moving in the forest, neither can one rely on technological superiority to overcome
today’s insurgents.
However, as any asymmetry in resources will obviously imply that one of the
forces will always have more power, the term “asymmetric warfare” is somewhat
misleading.8 The strong force would be supposed to win always or at least most
of the time against the weak one.9 History, though, often shows the opposite, as
discussed in Chap. 4. Rosen (1972) offers alternative explanations, implying that it
would be necessary to observe other sources of asymmetry, like the asymmetry in
willingness to absorb costs.10
One of the first articles explaining outcomes in asymmetric conflict is “Why Big
Nations Lose Small Wars” by Mack (1975). The case of Vietnam and the Algerian
war are used among other examples to show that superiority in military capability
does not necessarily mean victory in war. In the history of imperialistic expansion
we find local forces taking up arms against an external force, which is superior both
militarily and technologically. Furthermore, as the local insurgents do not have the
power to invade their opponent’s homeland they do not constitute a direct threat
to the external force’s sovereignty, either. Thus, while the war for the insurgents is
“total”, as the external force can threaten to invade and occupy its home territory,
it would not be total for the external force. It is this asymmetry that could make
the external force much more susceptible to the costs of the war, both in monetary
as well as human terms, as well as unable to mobilize all its resources because of
political reasons, giving rise to the differences in Rosen’s willingness to suffer costs.
For the insurgents, the one way left to win the conflict is to escalate it in terms of
both time and enemy casualties until the external force is no longer willing to absorb
such costs.
Asymmetries in the distribution of force are also decisive when it comes to
define the terms of the conflict. As the begin of an insurgency implies that insurgent
groups are dissatisfied with the status quo, the terms under which they will fight that
status quo will be defined by the resources they can command. Guerrilla warfare is
thus used by the insurgency, as it entails lower costs than conventional warfare. In
case guerrilla warfare becomes unaffordable, terrorism is likely to be the only way
left to unbalance a superior force, since terrorism, while not being able to yield
the same payoffs, would certainly exhibit lower costs. However, while resource
asymmetries can explain terrorism, they cannot be considered the origin for all
terrorist actions. Terrorism should be rather understood as part of a vicious circle

8
Additional information on asymmetry on resources is available e.g. in Katzenbach (1962).
9
Arreguín-Toft (2001) defines an actor to be strong if his material power exceeds that of his
adversary or adversaries by at least ten to one (Arreguín-Toft 2001, p. 94).
10
Yet, as discussed in Chap. 4 to Arreguín-Toft (2001) strong actors lose asymmetric conflicts when
they use different strategic interactions.
54 6 Irregular Warfare

of military actions and/or policies. Because of the threat of terrorism, and in order
to devise countermeasures, states tend to increase expenses in military technology,
thereby increasing the asymmetry with regard to the insurgents, and simultaneously
the risk of terrorism (see, e.g., Butler and Gates 2010 and Rosendorff and Sandler
2005).
Betts (2002) analyses the attacks which occurred on 11 September 2001 and
illustrates how power itself can be a source of vulnerability. 9/11 was a surprise
even for some national security experts. Back in 2001 it was hard to believe that
a country whose military budget was almost five times higher than the combined
military budgets of all potential enemy states could be attacked on its home soil:
There was a general perception in the US of war as being something that happened
abroad (see Betts 2002, pp. 22–27). But in the 9/11 attacks, a small number of
men killed 3,000 people and cost the US economy 1.8 million jobs (The New York
Times 2002). According to Betts (2002) “primacy could, paradoxically, remain both
the solution and the problem for a long time” (Betts 2002, p. 36).
Subsequently, US military expenditure steadily increased since 2001 to peak
in 2011, when it reached 4.6% of the GDP. It was only due to the August
2011 enactment of the Budget Control Act (BCA) that defense spending for the
fiscal years between 2012 and 2021 became severely limited: military expenditure
decreased to reach its lowest level in 2014 accounting for 3.5% of the GDP.11
However, this value is still high when compared to other countries. The combined
European military expenditure12 e.g., has since 2006 come down to account for
1.54% of the GDP in 2013, while the respective US share in that same year was
3.8% (European Defence Agency 2015).
For Mack (1975), the first condition for avoiding defeat in an asymmetric
conflict is to refuse to confront the enemy on his own terms. The insurgents often
rely on the mountains and forests and on the shelter of bordering countries to
protect them from being attacked, as it happened in Vietnam during the post-1968
North Vietnamese operations along the Ho Chi Minh Path, where insurgents were
sheltering in the neighboring countries of Laos and Cambodia, or in Afghanistan
where the Mujahidin guerrillas, today’s Taliban, were using mountainous terrain

11
The BCA caps do not affect funding for Overseas Contingency Operations (OCO). As the budget
for the fiscal year 2016 in national defense spending requested $561 billion, $38 billion above
the BCA defense cap, the current debate now is whether to adjust the BCA defense cap, move
base budget spending accounts to accounts for OCO to avoid the BCA cap, reduce the defense
expending or a combination of these to avoid the sequester (Belasco 2015).
12
There are some discrepancies about the European budget figures on defense. According to
European Defence Agency and NATO European budgets, the proportion of defense budgets in
European national GDP decreased between 2008 and 2010 (European Defence Agency 2013)
while according to the Institute for Security Studies’ (ISS) 2014 Military Balance, the European
defense budget rose from 1.45% to 1.58% (Institute for Security Studies 2014, p. 97). The
discrepancies are explained by whether military pensions are included in military budgets and
the countries included. ISS considers countries from a geographical definition of Europe while
EDA uses an institutional approach. We refer here to the European Union and therefore take the
data from the 27 EDA member states (all EU member states except Denmark).
6.2 Defining Irregular Warfare? Some Comments. . . 55

to ambush Soviet and, respectively, ISAF forces and/or just took shelter across the
Pakistan border. Conventional military forces, despite their superiority in resources,
were and still are not sufficiently well prepared or not allowed to intervene in the
first place.

6.2 Defining Irregular Warfare? Some Comments

on Objectives in Irregular Warfare

The very first problem arising when it comes to studying strategies on how to fight
in an irregular warfare situation is the fact that there is no universally accepted
definition of it (see Hurley et al. 2007). Understanding the nature of irregular warfare
is paramount for the force planner. While the ambiguity of the available definitions
has already been discussed, one further such definition is to be introduced here:
the “Institute for Defense Analyses” (IDA) defines irregular warfare by identifying
attributes that distinguish it from “regular warfare”13 and uses Iraq and Afghanistan
as case studies. The main difference is that the IDA is not content with merely
defining irregular warfare but at the same time tries to pinpoint those aspects of
irregular warfare that help find new ways of how to conduct COIN operations in an
irregular-warfare environment.
The final objective of every conflict is, of course, to win. However, in regular
warfare this is not just accomplished once the opposing force is defeated, but it
often requires the provision of institutions that enable a legitimate government.
Many armed forces have recently changed their doctrines to “population centric”
approaches where the civil population and the impact military campaigns have on
them is taken into consideration, understanding that the stability needed to build
institutions relies directly on the support thereof. The IDA e.g. argues that long-
term stability will not come from winning the “hearts and minds” of the people but
from helping the host nation government to win “the hearts and minds” of its own
people (Hurley et al. 2009).
The IDA distinguishes five attributes that need to be considered while approach-
ing irregular warfare (Hurley et al. 2007, p. II-2):
1. The central role of human terrain
2. Extraordinarily tight coupling of civilian and military operations and activities
3. Military actions emphasizing small units operating among civil population
4. The need for security, stability, and reconstruction to be consolidated within an
area, rather than just defeating the enemy and moving on, as in regular warfare
5. The overarching objective of transition to a secure, stable, legitimate host nation
government.

13
One definition of “regular warfare” i.e., “warfare between permanently organized bodies, each
constituting the standing force of a state” (Hurley et al. 2007, p. ES-1).
56 6 Irregular Warfare

Each of these attributes points at the crucial role that has to be attributed to the
population. Providing security is only possible with the host nations’ cooperation.
Once the ground has been taken from the insurgents, the next objective, in order
to provide security and stability, should be the establishment of a reliable local
police system without which no host nation’s government could hope for being ever
considered to become legitimate. Merely gaining terrain would thus be pointless
unless COIN forces are able to subsequently provide security for the civilian
population.
The objective of providing security and stability is twofold: on one side it aims
at winning support making it more difficult for insurgents to hide; on the other it
tries to put an end to the recruitment of new fighters into the insurgents’ ranks.
The willingness of the population to support one or the other force will depend
first on the ability each force has to assure their physical security, and, secondly, on
their closeness to their social and political preferences.14 Non-kinetic capabilities—
all capabilities relevant to irregular warfare that do not focus on destroying enemy
forces through the application of physical effects (Hurley et al. 2009, p. ES-1)—
have proven to deliver higher returns on population support.

6.3 Designing Operations in Counterinsurgency Warfare

In order to make every soldier understand better the ever changing nature of wars
they are likely to fight, and the challenges they are likely to face, armies continually
publish and refine respective doctrines and guidelines. In the US, the Department
of Defense, on top of releasing such documents for each of the different forces
to provide force commanders with current guidance for conducting, planning,
preparing, executing and assessing military operations, has turned to issuing so-
called “Joint Publications”, thereby stressing the importance of different forces,
such as, e.g., the Army and the Marine Corps, to embark on joint operations.15
A military operation is a set of actions intended to accomplish a task or a mission.
They can rank from stability operations to homeland defense. The need to act jointly
holds all the more so when it comes to COIN operations: JP 3-24 is, e.g., devoted
to counterinsurgency operations (Joint Chiefs of Staff 2013). Joint publications
provide military strategists and war analysts with a broad perspective on the aspects
considered in the design of military operations.
Hurley et al. (2007) use the example of Iraq to illustrate a so-called “Irregular
Warfare Generic Blue Mission Map”, which can easily be extended to many of

14
Hurley et al. (2007) divide the population into four categories depending on their preferences
related to who (Blue or Red) they believe is more important to their physical security and who they
believe better fits their socio/political/economic preferences.
15
The joint publication 3-0, e.g., stands for Operation Series, providing doctrinal foundation and
fundamental principles that guide the US armed Forces in the conduct of joint operations across
the range of military operations.
6.3 Designing Operations in Counterinsurgency Warfare 57

today’s other irregular warfare situations. It includes all missions in support of “Iraqi
Freedom” and lists all the capabilities that were considered necessary to accomplish
them.
In total, Hurley et al. (2007) counted 23 missions and 92 capabilities. Out of
the 23 missions only six had an unambigous military nature covering combat and
support tasks; they consisted of neutralizing Red forces, protecting Blue forces,
protecting physical sites, working with indigenous security organizations, standing
up new indigenous military units and protecting the population (Hurley et al. 2007,
p. IV-4).
When it comes to the design of military operations, the force commander is well
aware that the operations’ objectives are conditioned by the character of the conflict.
Operations in irregular warfare are usually characterized by the low intensity of
its operations in order to minimize the footprint left on the host nation. It is often
implicitly assumed that, while the superior force’s primary objective would be to
achieve victory, the weaker force would rather aim to hold as long as possible in
order to exhaust the bigger force; thus, due to the imbalance of power between
the forces, the weaker force usually does not have other alternatives than to defeat
their enemy by political exhaustion. Then, to reduce the political side effects,
operations in irregular warfare would favor the use of light infantry tactics and
special forces. Methods usually employed to defeat insurgents include attrition,
decapitation, i.e. going after local commanders, intelligence raids and ambushes
to attack the recruitment, funding and supply of the insurgents.
As mentioned before, one of the aspects distinguishing irregular warfare from
conventional warfare, is the way battles are conducted, and in particular the use of air
power. The idea of using aircraft in direct support of ground troops goes back to Italy
and the Italo-Turkish War of 1911. In 1917, following the recommendation of and
authorization by the Commandant of the United States Marine Corps, Major General
William Biddle was tasked with establishing a Marine Corps aviation company, and
since then the idea of Marine Aviation to be used as direct tactical support of Marine
Corps infantry has taken hold firmly (Alexander 2012). It was employed for the
first time in the so-called Banana Wars when US Marines and Nicaraguan police,
surrounded by revolutionary troops, called for air support using large panels laid out
on the ground.
At present, the air dimension is essential for tasks like reconnaissance and
surveillance, for the troop mobility and of course for conducting CAS. Drawing
from a definition, CAS covers all “air action by fixed- and rotary-wing aircraft
against hostile targets that are in close proximity to friendly forces and require
detailed integration of each mission with the fire and movement of those forces”
(Joint Chiefs of Staff 2015, p. 34).
The integration of CAS into Marine Doctrine and its role as a tactical, rather than
strategic means, was most clearly emphasized by Lieutenant General Victor Krulak,
who stated in 1968: “Air support is as inseparable to the combat team as his artillery,
his tanks, or even the infantryman’s M-16” (Shulimson et al. 1997, p. 501). CAS has
been a standard procedure in Iraq, and Afghanistan.
58 6 Irregular Warfare

In summary, COIN operations would often consist of missions aiming to liberate

areas occupied by insurgents, where air capabilities play a key role, and the resulting
casualties on friendly forces and on civil population will significantly shape the
future terms of the conflict. In addition, it will also be required to assure military
presence on the ground to support a solid transition to a legitimate host nation
government.
Chapter 7
Modelling Specific Aspects of COIN Operations:
On the Possibility of a First Mover Advantage

7.1 “The Surge” Versus a Surge

“Mission Accomplished” said the banner on the aircraft carrier Abraham Lincoln
from which in May 2003 the then US President, George Bush, declared the
conclusion of the military operations against Iraq. Nothing could have been further
from the truth, which was highlighted when, in the fall of 2006, the security
deterioration in Iraq peaked.
Irrespective of all the claims and counterclaims regarding the existence of
so-called weapons of mass destruction, the way operation “Iraqi Freedom” was
conducted had been controversial almost from the start. The “shock and awe”
campaign began on March 20th, 2013. It aimed at toppling Saddam Hussein’s
government and was presumed to be over in a couple of months if not weeks. It
turned into a war lasting until 2011, with over 4,000 American soldiers dead (Fischer
2015, p. 1), close to half a million Iraqi casualties (Hagopian et al. 2013) and costing
815 billion dollars.1
While the forces deployed in 2003 were large enough to successfully remove
Saddam Hussein from power—the regime collapsed in April 2003 as Baghdad came
under American control—, they were insufficient to provide security to the Iraqi
people in the aftermath of the invasion. What followed proved how ill-prepared the
US Army was to fight an enemy in an irregular warfare context.

1
These are the costs estimated for both OEF (Operation Iraqi Freedom) and OND (Operation New
Dawn) (Belasco 2014). Bilmes and Stigliz (2008) estimated the economic costs going forward the
Iraq and Afghanistan war would reach the three trillion dollars. Later, Bilmes and Stigliz (2011)
concluded that the costs going forward would surpass their first estimations. Their estimations are
built taking into account economic costs including medical care, paying disability compensations
for Veterans, the costs of the escalation war in Afghanistan and the ones attributed to the Iraq War
part of causing the Great Recession.

© Springer International Publishing AG 2017 59

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_7
60 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

Fig. 7.1 Security in Iraq, January 2004–March 2010 (Detail, Special Inspector General for Iraq
Reconstruction 2010, p. 49)

When the US transferred the formal sovereignty of Iraq to its new leaders after
little more than a year of occupation, the deterioration of security was already
evident.2 Four months later, important areas of central Iraq were in the hands of
the insurgents.
By the end of 2004, Iraq was in the middle of a spread of sectarian violence.
Sunni groups resisted the establishment of the new government by launching indis-
criminate IEDs and suicide attacks to intimidate potential voters and delegitimize
the electoral process. Official Iraqi Security forces were continuously conducting
targeted raids and death squad activities against Sunnis. What at first appeared to
be mere acts of hostility between Sunni and Shiite neighbors, proved to be a fight
for dominance between rival terrorist groups. As Fig. 7.1 illustrates, the violence
reached its highest level after 3 years of occupation.3
The events that followed the occupation of Iraq in 2003—the escalation of
violence that put the country on the verge of a civil war and the Americans’ loss
of control over many cities which quickly ended up in the hands of the insurgents—
was a direct consequence of the misconception the US had on the conduct of a
conflict like the one in Iraq. There are two aspects that, according to many theorists,
may explain what went wrong when what was designed to be a small and quick
invasion for the US became the most deadly conflict since Vietnam.
One aspect was the policy of then US Secretary of Defense, Donald Rumsfeld,
to win the war on the cheap by turning down the US Army’s request for more troops
long before the invasion began. The other was America’s misconception that this
war basically was a kind of re-run of the first Gulf War: while the 1991 conflict was
in a way similar to the kind of war the US had been preparing for decades during the

2
Just less than 1 month before, the ambush of a convoy from the private military company
“Blackwater USA” shocked the public opinion were the images of the event were released by
the press, showing Iraqis posing with the corpses before being hung over a bridge crossing the
Euphrates river.
3
See, e.g., the Sadrist movement (Cochrane 2009).
7.1 “The Surge” Versus a Surge 61

Cold War—the tactics of tank warfare that would have been employed in order to
defeat a Soviet invasion in the Fulda Gap were the same that were used to defeat the
Iraq Imperial Guards—, this one would turn out to become irregular. The majority
of US forces deployed had hardly any experience in counterinsurgency operations.
The counterinsurgency doctrine they relied on in 2003 was more than 20 years old,
and it was largely forgotten, too. The first mentioning of this new kind of warfare
today described as irregular warfare goes back to then President John F. Kennedy
in his speech to the Graduating Class of the West Point Military Academy on June
6th, 1962 in which he foresaw the kind of problems to be faced by the US in the
Vietnam War (Kennedy 1962). Dealing with the concept of irregular warfare would
thus only have reminded the US military of one of the most traumatic conflicts in
its history; ignoring it was an easy option. Unfortunately, it deprived the American
forces of 2003 of the knowledge how to fight the next irregular war.
Insurgents could easily flee from the hot spots and would just wait for the US
forces to retreat once “victory had been accomplished” only to come back and regain
control with minimal effort. It not only destabilized the country but also provided
insurgents with a popular support based on fear. The small amount of US troops
deployed in Iraq made the Iraqi population aware of the imminent return of the
insurgents. It should not come as a surprise that people refused to collaborate with a
force that was not going to be present any more once the insurgents returned. Kagan
(2009, p. 5) remarked that the story of Al-Qaeda in Iraq during 2005 was “[T]he
story of its dispersion from Fallujah and its reestablishment elsewhere in Iraq”.
As has been discussed in the previous chapter, both aspects are but two sides of
the same coin, as success in counterinsurgency critically depends on how well the
population can be protected. Such a task, though, becomes impossible to succeed
in without a sufficient amount of forces on the ground (Department of the Army
2014). The troop density recommended by Quinlivan (1995), a senior operations
research analyst at the RAND Corporation, which is 20 to 25 counterinsurgents
for every 1000 residents, has become conventionally accepted. Quinlivan (1995)
actually suggested different ratios depending on the level of violence and threat
intensity. A ratio of 20:1000—as seen in Malaya (1948–1960) and in Northern
Ireland (1969–2007), when COIN forces found themselves in situations where the
rule of law had collapsed and operations had to be designed to maintain order—
would be placed on the high end of the scale (Quinlivan 1995).
Condolezza Rice, the Secretary of State at the time, was one of the first voices
announcing how the new strategy employed in Iraq should look. On 19 October
2005, on the Opening Remarks Before the Senate Foreign Relations Committee,
she opened her speech referring to a different approach to face the war in Iraq:
I’ve spoken many times about why we are there, but I would like to talk about how we
assure victory. In short, with the Iraqi Government, our political-military strategy has to be
to clear, hold, and build: to clear areas from insurgent control, to hold them securely, and to
build durable, national Iraqi institutions (Rice 2005).

A year later, on 15 December 2006, the Counterinsurgency Field Manual (FM

3-24) was released (Department of the Army 2006). It emphasized the primary role
62 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

of non-military activities over military actions, which should be understood merely

as actions subordinated to political ones. The central idea defended in the FM 3-24
is that the center of gravity in an insurgency—expressed in Clausewitzian terms—is
the population. Once it comes to an insurgency, the population would have to decide
whether to support the insurgents or the counterinsurgency forces. The more support
either side gets, the more likely it will achieve victory. At the same time, it will
influence the degree of legitimacy of the host nations government, being essential
for economic and institutional development. The message conveyed by FM 3-24 is
clear: unless the host nation is capable to provide security and stability, there is no
way the insurgency can be defeated.
In terms of execution, the FM 3-24 summarizes counterinsurgency in three lines
of action. The first one would be to “clear, hold and build”. The other two involve
a close cooperation with the host nation; one using different actions combined with
the host nation’s troops and the other trying to maintain a minimal as possible
presence in the country while at the same time providing the necessary support.
The approach taken in the FM 3-24 is also known as population-centric; it can be
understood as opposed to the enemy-centric approach. The enemy-centric approach
states that it is the insurgents’ defeat that will cause their collapse. Examples of
these two approaches are found in the shift of the strategies followed first by the
British counterinsurgency in Malaya and similarly in the US Army strategies in
Iraq. In Malaya, the two British Generals in charge of operations found themselves
at the opposite ends of this spectrum: while Harold Briggs chose the enemy-centric
approach and conducted operations using search and destroy techniques to kill
the insurgents, his replacement, Sir Gerald Templer, went for a population-centric
approach and considered the support of the population to be crucial and therefore
employed strategies to win the “hearts and minds” of the population.4 The change
in Iraq was brought about by General David Petraeus: he not only co-authored FM
3-24 but he implemented the new counterinsurgency campaign when he became
Commanding General of the Multi-National Force Iraq in February 2007.
However, there are many theorists like US Army Col. Gian Gentile or the
British military theorist William F. Owen, who criticize the suitability of the
population-centric approach. Their critique relies on the strong conviction that a
counterinsurgency can only be successful by seeking and destroying the enemy, i.e.
the belief that population-centric methods are of little benefit to fight insurgency.
Gentile (2009) directly attacks the population-centric approach emanating from
the FM 3-24. He considers that the Field Manual contains no strategies or
methodologies other than the winning of the “hearts and minds”, and even though
this may be considered as a reasonable operational method in certain circumstances,

4
The FM 3-24 often refers to the “lessons learned” in Malaya, assuming the “winning of hearts
and minds” together with the British Army as a “learning organization” was what finally ended up
with the insurgency. Nevertheless, authors like Hack (2009) argue that the credit of this strategy
to explain the victory may be exaggerated, as the high-level insurgency was already broken—
incapable of turning the advantage of the British forces around—in 1950-1952, while the winning
of hearts and minds was still in an early stage (Hack 2009, p. 384).
7.1 “The Surge” Versus a Surge 63

it cannot be seen as the only means to fight against insurgency, since history has
shown cases where insurgencies were defeated in ways that have little in common
with the population-centric approach.5
Likewise, Owen (2011) criticizes the people-centric British joint publication on
security and stabilization—“British Joint Doctrine Publication, JDP3-40: security
and stabilization: the military contribution”—arguing that the way to win in
irregular warfare does not differ much from the way regular forces are defeated.
For Owen (2011) the people’s protection should not be the responsibility of the
army, but a benefit gained by the population when the army destroys the enemy.
Responsibility for protecting the population should lie with law and order forces
rather than the army. In Owen’s view, the source of confusion comes from the idea
that supporting the population creates power, when to him it is power that creates
support (Owen 2011, p. 36).

7.1.1 “The Surge”

Following the new US Army doctrine, the population-centric strategy planned for
Iraq in 2007 required the provision of security which, as explained before, could
only be achieved with ground forces. Consequently, the US decided to increase
their deployment in Iraq by sending five additional US Army combat brigades in
what later became known as “The Surge”. In less than a year the security in Iraq
improved to levels comparable to those from the first year of the conflict.
Once “The Surge” moved into the conflict areas and established itself there, it
should become easier to make the population believe in their commitment to assure
security in the area. In that way the Iraqi population would no longer have to fear
the return of the insurgents, resulting in switching their support to the US forces and
encouraging them to start fighting the insurgency themselves. “The Surge” should
thus have caused a domino effect: the increase of US troops improved security,
which in turn increased the number of Iraqis supporting the stabilization process
(see, e.g., Downey et al. 2008).
There are alternative explanations for the dramatic fall of violence in Iraq during
the second half of 2007. One of them is that, by the time the Surge was deployed in
Iraq, the sectarian bloodshed was for the most part already over.6 Others argue that it
was actually the Sunni tribal uprising against Al-Qaeda, which produced the “Anbar
Awakening” in 2006 and 2007, that deserves the most credit for putting an end to
the spiral of violence in Iraq. Biddle et al. (2012) show evidence suggesting that
it was the synergy between “The Surge” and the “Awakening” that would explain

5
“Currently, US military strategy is really nothing more than a bunch of COIN principles, massaged
into catchy commander’s talking points for the media, emphasizing winning the hearts and minds
and shielding civilians. The result is a strategy of tactics and principles” (Gentile 2009, p. 15).
6
“[T]he killing stopped because there was no one left to kill” (Cockburn 2008, p. 14).
64 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

better why the violence decreased in Iraq in 2007. They believe that “The Surge”,
understood not only as an increase of the number of troops on the ground, but also
as troops employing a brand new doctrine in counterinsurgency, was only fit to put
a temporary end to violence, but could hardly break the insurgency.7

7.1.2 The Model

Let us assume the scenario described above: The escalation of violence experienced
in Iraq in 2006 forces the US Army to consider a decisive increase in the amount
of ground troops to fight the insurgency. The alternative, more in line with Donald
Rumsfeld’s policy of facing a conflict with a minimum amount of forces deployed,
would only work to maintain the status quo. We distinguish between two possible
strategies for the US: SURGE for the case it sends extra ground troops to Iraq and
STATUS-QUO otherwise.
The insurgency, represented here by Al-Qaeda Iraq (AQI), will also have to
decide whether to continue fighting, a strategy we will call UPRISING or whether
they rather remain inactive for a time, only postponing the uprising until the US
forces leave, which we will call INACTIVE, yielding a payoff matrix (Fig. 7.2).
We assume the structure of all possible outcomes in such a scenario to be as
follows: The US would prefer any situation where Al-Qaeda remains inactive, and
would prefer this to happen without having to deploy any additional military units
in Iraq. In case an uprising takes place, the US would be better off by having as
many troops as possible, since it would increase the likelihood of a quick defeat of
the insurgents. The payoff’s distribution would then be given by:

B4 > B2 > B1 > B3 . (7.1)

On the other hand, Al-Qaeda has the ultimate goal of expanding its dominance
throughout the country. It is obvious that the less ground troops are available to

Fig. 7.2 Payoff matrix for

“The Surge”

7
For a statistical and geospatial analysis about the connection between increases on troops and
increased security in Iraq, see Thiel (2011).
7.1 “The Surge” Versus a Surge 65

the US, the greater the chances they succeed in maintaining their hold over the
territory. However, given the asymmetry of resources between the US Army and the
insurgents, once the US plays a SURGE strategy against UPRISING, the expected
outcome will be a sound victory for the US which, at the same time, will mean
the end of the insurgency, and therefore will provide the insurgents with their worst
possible outcome. In case the US is deploying the SURGE, the insurgents will rather
prefer to move to other areas out of reach of US forces and wait until the US forces
leave before launching any attack. If they postpone the uprising and remain inactive,
they would prefer the US to deploy the SURGE, because it would not only give rise
to opportunity costs for the US government but it would also work as a call for new
recruitments to their ranks because of the US troops’ larger footprint. The payoff’s
distribution will thus be given by

R3 > R2 > R4 > R1 . (7.2)

As there is no Nash equilibrium in pure strategies, a Nash equilibrium in mixed

strategies exists. Then, the US will play SURGE with some probability p while AQI
will play UPRISING with some probability q.
There would be an alternative, though. In theory, either player could opt to
commit to one of its strategies beforehand. In practice, though, as there is no way
AQI could be made to commit to any strategy, due to a respective lack of institutions,
implying that any such attempt would not be subgame-perfect, such a “first move”-
strategy could only be employed by the US. Assume now that the US pre-announces
and/or deploys a SURGE before AQI can make its move. Then the insurgents would
respond with INACTIVE, resulting in a payoff of B2 for the US forces. Comparing
this to the expected payoff resulting from the mixed-strategy equilibrium reveals
US
ESURGE D B1  q C B2  .1  q/ < B2 (7.3)

because of B1 < B2 . Thus, if from the very beginning the US would have been in
the position of deciding whether to deploy a large amount of troops or not, they
would have done so. Comparing all possible alternatives we find Table 7.1 showing
that there is a second mover advantage, as anyone who succeeds in moving second
would enjoy a higher payoff. However, as mentioned above, only AQI can play
second.
In addition, if we assume the US uses the strategy that minimizes its losses in
view of their enemy’s capabilities (i.e. max-min strategies), we find that SURGE
is once again the best strategy to follow. In case the insurgents follow the same

Table 7.1 “The Surge”. Simultaneous First mover Second mover

Expected outcomes
US B1 < E.B/ < B2 B2 B4
AQI R4 < E.R/ < R2 R4 R2
66 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

approach—to minimize their possible losses—they will choose INACTIVE, all of

this resulting in the same equilibrium obtained when the game is played sequentially
with the US moving first.
It may therefore be concluded that deploying the SURGE from the very
beginning would not only have been the best strategy for the US, but also the one
where the maximum losses for each side was minimized.

7.2 “Clear Hold and Build” vs “Search and Destroy”

The two features characterizing “The Surge” were the large amount of forces
deployed on the ground and the change to a brand new counterinsurgency doctrine.
The new doctrine, as discussed before, was focused on the people’s protection,
following the belief that population support was the only way of overthrowing
insurgencies.
The strategy, doctrinally established in the FM 3-24, is tactically divided into
three steps: clear, hold and build. In the 2014 version of the FM 3-24, though, “Clear,
Hold and Build” has even been expanded to “Shape-Clear-Hold-Build-Transition”
(Department of the Army 2014).
The strategy “Clear, Hold and Build” rests on the idea that no insurgency
can survive without the population’s support. One common problem to fight an
insurgency is that, while COIN forces need to be continuously alert and ready to
fight insurgents everywhere, an insurgency is free to choose the target and the time
to strike. This recalls the old COIN aphorism that says that the insurgency is winning
if it is not losing.8 If, on the contrary, it is the COIN force that has the population’s
support, this can pay off in the ability of collecting more intelligence, up to the point
where the insurgents would have nowhere left to hide.
The effectiveness of “Clear, Hold and Build”, if compared to Vietnam-style
“Search and Destroy”, lies in its sustainability. While a “Search and Destroy”
strategy can literally kill the insurgents, it may not be able to beat the insurgency.
Thus, it can even have the side effect of encouraging the recruitment of new adepts
for the cause. A “Clear, Hold and Build” strategy on the other hand seeks out a
long-lasting solution by building a stable system where there would be no room left
for the insurgency.
The actions of the three actors involved in a “Clear, Hold and Build” strategy can
be described as shown in Table 7.2.
In short, a “Clear, Hold and Build” strategy aims to push the insurgents out,
thwart any possible return, and build institutions stable enough to promote economic
growth and social development.
The first two steps, “Clear and Hold”, are of evident military nature. The third
one, on the other hand, requires non-military assets and is of such a complexity that

8
“The guerrilla wins if he does not lose” (Kissinger 1969, p. 214).
7.2 “Clear Hold and Build” vs “Search and Destroy” 67

Table 7.2 Actions in “Clear, 1. Clear 2. Hold 3. Build

Hold and Build”
Insurgents Hold Inactive Inactive
COIN Offensive Defensive Stability
Host nation Hostile Neutral Positive

undertaking it is perhaps the real challenge of a people-centric counterinsurgency’s

approach. Jeffrey (2015)—the US Ambassador to Iraq from 2010 to 2012—argues
that it is in this last step, “Build”, where the Counterinsurgency strategy failed.
He argues that building institutions in the host nation requires firstly local forces
capable of replacing US forces and secondly the presence of US ground troops
throughout the entire process. A continuing presence of ground troops, though, not
only runs the risk of fuelling negative reactions from the host nation’s population,
as it could be regarded as a neocolonialist intrusion, but could give rise to mistrust
in the neighboring countries who would rather prefer the insurgency to win, rather
than see the supposedly “colonial power” establish a base on their footstep.9 During
the time he served in Baghdad, Jeffrey (2015) observed how most critical reforms
usually worked only under constant US supervision. In his view, what happened in
Iraq in 2011 with strongholds falling into the hands of Al-Qaeda shortly after the
US troops left was inevitable. Jeffrey (2015) remarks “If no real threat appears, the
state stumbles on; if one emerges, as the Islamic State of Iraq and al-Sham (ISIS)
did in 2014, disaster follows” (Jeffrey 2015, p. 178).
After the implementation of “Clear, Hold and Build” in Iraq, the US forces
sought to apply the same strategy in Afghanistan. By the end of 2001 the US had
successfully defeated Al-Qaeda and the Taliban, and the number of foreign troops in
Afghanistan and the resources for the country’s reconstruction were therefore min-
imized.10 The country remained quiet until 2006 when the Taliban regrouped and
began to launch attacks from their traditional strongholds in Kandahar and Helmand
as well as from neighboring Pakistan to retake Afghanistan from NATO. The Taliban
had managed to create a “shadow government” and succeeded by extending their
influence upon the population who had no connection with the NATO forces. In June
2009 the US conducted a revision on their strategy in Afghanistan that resulted in
the recommendation of a population-centric counterinsurgency effort; in November
2009 a troop increase of 30,000 focusing on the population was approved (Friedman
2014, p. 89). However, such a strategy would not work as it did in Iraq which might
be explained by three sociocultural factors that vary widely from one country to the
other. While in Iraq the population is concentrated in cities, in Afghanistan 77% of
the population lives in rural areas. Conducting “Clear and Hold” operations in rural

9
As observed e.g. with China and North Vietnam in South Vietnam or Syria and Iran in Iraq.
10
The amount of troops in Afghanistan back then was the smallest US keeping force since World
War II and the aid per capita much lower than the one received by Bosnia in the 1990s (Friedman
2014, p. 87).
68 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

areas requires naturally more resources and troops. Furthermore, the social structure
in Afghanistan is still tied to strong tribal loyalties; and the fact that Afghanistan is
one of the countries with the lowest literacy rates in the world represents a natural
barrier when trying to win the “hearts and minds” of the population. The weak
central government and its large and porous border also contribute to hindering the
efforts of carrying out any “Clear, Hold and Build” strategy as was conducted in
Iraq.
Finally, the costs of “Clear, Hold and Build” are also to be taken into consid-
eration. A significant part of the services necessary to “Clear, Hold and Build”
are provided by contractors. In 2008 the estimated ratio of contractor to military
personnel presence in US military operations in Iraq was reported to be about one
to one, a value only matched by the Balkan conflict and significantly higher than in
any other registered military operation. The DoD policy in Iraq assumed the military
provision of security to contractors and government civilians only if they were
deployed with a combat force or if they were directly supporting a military mission.
Then, all US agencies and contractors such as reconstruction contractors that did
not fulfill this requirement required the provision of their own security, resulting in a
considerable increase in private security contractors (PSCs) (Congressional Budget
Office 2008, p. 13). In addition, there are other areas like logistic support, equipment
or training which are also responsible for the substantial increase in the dependence
on contractor support.11

7.2.1 The Model

In the context of the above-mentioned spiral of violence in Iraq, we will analyze the
implications of employing the new US counterinsurgency doctrine, i.e. the people-
centric approach characterized by the use of a “Clear Hold and Build” strategy
(CHB), instead of the enemy-centric approach represented by the “Search and
Destroy” (S&D) strategy used in, e.g., Vietnam. The insurgency, in turn, has to
decide whether to play UPRISING or whether to withdraw (W/DRAW).
The payoff matrix would be given in Fig. 7.3.
Regarding the COIN forces’ preferences, the main objective will be accom-
plished when the insurgents retreat. Because a CHB strategy requires more troops
and more resources when compared to S&D, the US forces would prefer if the
insurgents would retreat on these terms, meaning B4 > B2 . In case of an uprising,
the US forces would be better off by having a larger presence on the ground and
implementing a CHB strategy which would be expected to deliver a long lasting

11
Additional information on the costs of contractors in Iraq and Afghanistan is available from
Markowski and Hall (2011).
7.2 “Clear Hold and Build” vs “Search and Destroy” 69

Fig. 7.3 Payoff matrix for

“Clear, Hold and Build”

solution to the uprising, all of this resulting in

B4 > B2 > B1 > B3 .12 (7.4)

Turning to the insurgents, it can be assumed that they would assess the conse-
quences of a CHB strategy as the definitive loss of their influence in the disputed
area. Therefore, the payoffs resulting from CHB would definitely be rated as the
worst possible scenario for the insurgents. Still, in case the COIN forces use a
CHB strategy, they would prefer to withdraw if only to have a chance to live for
another day since a successful uprising would be nearly impossible, hence R2 > R1 .
In case the COIN forces play S&D, the insurgents would be better off by playing
UPRISING.
There are three factors that explain these preferences; first we should not forget
that the insurgent’s objective is to gain control of the area. Even though they
hardly have any chance against a militarily superior COIN force and thus could
not reasonably aim for a victory in absolute terms, they would be perfectly happy to
perpetuate the conflict for as long as possible because, as has been seen before, “the
insurgence is winning if it is not losing”. Secondly, an unequal struggle would help
to lure new volunteers into the ranks of the insurgency. And thirdly, the insurgents
are aware of the pressure the public opinion has on the deployment of military forces
abroad and that an ongoing “Search and Destroy” strategy can easily erode public
support for the counterinsurgency and even result in the withdrawal of the troops
deployed in the host nation. The payoff’s distribution will then be given by

R3 > R4 > R2 > R1 . (7.5)

Again, there is no Nash equilibrium in pure strategies. The equilibrium in mixed

strategies tells us that the COIN forces would be playing CHB with a probability
R3
equal to R1 RR24R 3 CR4
, while the insurgents would play UPRISING with a probability
B4 B2
equal to B1 B2 B3 CB4 , resulting in expected payoffs which for both actors are better

12
Note that this payoff distribution is identical to the one discussed in the previous model.
70 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

Table 7.3 “Clear, Hold and Simultaneous First mover Second mover
Build”. Expected outcomes
COIN B1 < E.B/ < B2 B2 B4
AQI R2 < E.R/ < R4 R4 R2

than their respective second worst but worse than their respective second best
outcome.13
The picture changes, however, in a game similar to the previous one, if
the COIN forces become the first mover by adopting a new counterinsurgency
doctrine: Even though effectively making a public announcement to switch to CHB
seems like handing an advantage to the insurgents by making the COIN strategy
more predictable, the resulting rollback equilibrium becomes (CHB)/(W/DRAW),
implying payoffs of B2 and R2 , thus making the COIN forces better off compared to
the simultaneous version of the game (Table 7.3).
On top of that, the insurgents as the second mover will achieve a payoff lower
than the expected payoff they receive by mixing their strategies in the simultaneous
version of the game. Theoretically they also could deviate from this outcome by
anticipating the COIN forces and moving first. In doing so, they would achieve a
payoff not only better than the outcome from the simultaneous and the second mover
scenario, but also better for the COIN forces, who would receive their best possible
outcome, and would therefore be happy by letting the insurgents move ahead.
However, as described in Sect. 7.1.2 on page 65, this is not realistically possible.
A strategic move happens only if, once it is done, there is no possible way to change
it. The COIN forces are able to move forward because, once the new doctrine is
implemented, doing otherwise is no longer possible. The insurgents, on the other
hand, are not able to act likewise. While, naturally, they are free to announce, for
instance, that they are withdrawing, it is impossible to be absolutely certain that
they will actually do that. Then, after announcing a withdrawal, if the COIN forces
react by saving the costs of “Clear, Hold and Build”, the insurgents would have an
incentive to launch an uprising instead of withdrawing, signifying that the COIN
forces should never give credit to the insurgents’ commitments.
If, on the other hand, we assume again that the US use the strategy that minimizes
their losses in view of their enemy’s capabilities (i.e. max-min strategies) we find
that CHB is—as in the sequential scenario—the best strategy to follow. If the
insurgents follow the same approach, respectively, they will choose W/DRAW. This
equilibrium does not differ from the one in which the COIN forces move first by
announcing a new counterinsurgency doctrine.
This works for the US on two levels: Firstly it is the risk-averse solution,
and secondly, it offers a sustainable solution to the conflict. In the event that the

13
As the payoff distribution for the COIN forces is identical to the one in the previous model,
its expected payoff by mixing the strategies will again never be higher than B2 . Likewise, for the
AQI
Insurgents; EW=DRAW D p  R2 C .1  p/  R4 < R4 ; because R2 < R4 .
7.3 “Clear, Hold and Build” and “The Surge” 71

insurgents consider a new uprising, the necessary institutions to deal with it would
already be in place. This solution also matches the strategy suggested in the former
scenario supporting the deployment of “the Surge”. This, together with the new
doctrine, constituted McChristal’s strategy to succeed in Iraq: “Protecting people
is the mission. The conflict will be won by persuading the population, not by
destroying the enemy” (Hall and McChrystal 2009, p. 1).

7.3 “Clear, Hold and Build” and “The Surge”

Whereas the implementation of the US forces’ strategies in the first model—

consisting of deploying “the Surge” or not—can be well isolated from the strategies
of the second model, it seems difficult to do the same the other way around and
separate the new COIN doctrine from “The Surge”. A “Clear, Hold and Build”
strategy could never be executed without the corresponding amount of troops on
the ground. Moreover, “The Surge” was, as mentioned before, deployed to procure
security and was therefore in line with the above discussed people-centric approach.
If we consider that there is an inextricable link between the COIN people-centric
approach and “The Surge”, it may be necessary to consider them together as a single
strategy.
While the COIN forces’ preferences remain the same—the payoff distribution is
equivalent to the previous two models—the insurgents have different advantages by
considering the outcomes of the “The Surge” and “Clear Hold and Build” separately.
Let us assume that, given the described spiral of violence Iraq was in during 2006,
the US are in the process of discussing the deployment of “The Surge” and assessing
whether to change their COIN doctrine from an enemy-centric to a people-centric
approach. Depending on the approach, if “The Surge” is approved, it could be
implemented together with the new COIN doctrine or without. Another possibility
would be to neither approve “The Surge” nor change the doctrine, in which case the
COIN would continue using “Search and Destroy” without any extra deployment of
forces.
The payoff distribution would be in line with the one described in the models
above. As the distribution of outcomes is different depending on the implementation
of the new doctrine, we will use the upper index d for the payoffs related to the use
of the new doctrine.
Let us assume that the largest fraction of the cost of implementing a people-
centric doctrine follows from the extra deployment of forces, making the cost
difference between a surge and a surge combined with “Clear, Hold and Build”
close to zero.
However, from the point of view of stability, a strategy that builds institutions
would achieve a long lasting solution to the insurgency. Therefore, the COIN forces
would prefer the outcomes where institutions are built. We will expect then the
outcomes related to the implementation of the new doctrine to be greater than the
enemy-centric approaches. Under these assumptions, the payoff distribution for the
72 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

Fig. 7.4 Payoff matrix for

“Clear, Hold and Build” C
“The Surge”

US will then be given by

Bd2 > Bd1 > B4 > B2 > B1 > B3 . (7.6)

The insurgents, on the other hand, will be worse off if the US change the COIN
doctrine. We will assume their payoffs after the implementation of the new doctrine
to be always worse than the ones with the enemy-centric approach. The payoff
distribution will then be

R3 > R2 > R4 > R1 > Rd2 > Rd1 . (7.7)

The resulting payoff matrix would be given in Fig. 7.4.

We find the COIN forces have a dominant strategy, namely deploying the Surge
together with “Clear, Hold and Build”. There is also a single Nash equilibrium in
pure strategies where the COIN forces deploy “the Surge” in the framework of a
people-centric doctrine and the insurgents withdraw.
The case at hand can again be modelled as a sequential move game, where the US
move first by deciding the doctrine and the strategy to follow, while the insurgency
reacts accordingly. Through backward induction it can be shown that this scenario
will deliver the same equilibrium.

7.4 Surge: An Enemy-Centric Approach

The Surge was deployed in Iraq together with a doctrine’s shift to a population-
centric approach. However, as discussed before, the deployment of a large amount
of forces is not necessarily used together with a change in the doctrine and could
therefore be implemented within an enemy-centric doctrine where attrition alone is
used to defeat the enemy.
7.4 Surge: An Enemy-Centric Approach 73

Fig. 7.5 Payoff matrix for

the enemy-centric surge

When a conflict with features such as the ones observed in irregular warfare
breaks out, the strong force is in the position of deciding the force intensity to
use against the weak force. The deployment of an overwhelming amount of troops
would leave the weak force no other option than to withdraw. On the other hand,
when COIN forces fight an insurgency with a more conventional amount of troops,
the victory of the strong force, in spite of the still asymmetric distribution of forces,
is neither evident nor immediate. The insurgency will find that, depending on the
size of the COIN troops, it could be wiser to withdraw and avoid a confrontation
while waiting for the conditions to become more favourable.
The difference to the model introduced in Sect. 7.1.2 lies in the final objective
of the COIN forces. In Sect. 7.1.2, the US forces’ main task was the provision of
security, the model discussed here seeks to determine the best strategies the COIN
forces have to end the insurgency, which, following the enemy-centric doctrine,
should be composed of strategies aiming for literally killing the enemy.
The COIN forces will either rely on a large deployment (SURGE), or will have
assigned an estimated minimal amount of ground troops which should be enough to
fight the insurgency, in which case only small raids against the insurgents (RAID)
can be conducted. On the other hand, the insurgents can choose between fighting
back (FIGHT) and a withdrawal (W/DRAW) (Fig. 7.5).
There is also a fifth option B5 =R5 , consisting of renouncing to launch any
operation at all.
Contrary to what we have observed in Sect. 7.1.2, the COIN forces would now
only succeed in their mission if they have assigned a large enough amount of troops
to successfully fight, i.e. kill, the insurgency, resulting in a sound victory for the
COIN forces. In case the insurgents withdraw, it becomes better to only have a
modest deployment on the ground as with no prospect of any fighting going on,
the opportunity costs of deploying a large number of troops could be avoided, thus
B4 > B2 . Then, the best outcomes for the COIN forces are now no longer the ones
in absence of confrontation but will depend on the insurgents’ strategies.
Should the COIN forces be deployed in a modest size, then the casualties they can
expect from fighting the insurgents will be considerable. In this case, renouncing to
launch an operation could prove a better alternative, as it would spare the casualties
expected from a confrontation in such terms. The payoff distribution for the COIN
forces would be given by

B1 > B4 > B2 > B5 > B3 . (7.8)

74 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

The insurgents are also pursuing their ultimate goal of defeating the COIN forces,
and this can only happen by fighting them. However, they only have a chance to
cause casualties if the amount of the COIN forces deployed is small. Thus, if they do
not meet the COIN forces in RAID, they would prefer the absence of an operation.
Otherwise, they will be better off by withdrawing, causing their preferences to
remain the same as in the model from Sect. 7.1.2.
The Insurgents’ payoffs will then be given by

R3 > R5 > R2 > R4 > R1 . (7.9)

As can be seen in the payoff’s distributions, the game presented here, contrary to
the model from Sect. 7.1.2, is a zero-sum game.

7.4.1 Simultaneous Game

We assume that the insurgents are on the verge of suffering an imminent attack from
the COIN forces, but at the time they have to decide whether to fight or withdraw,
there is no way to know the kind of troops they would meet on the battlefield.
Furthermore, the COIN forces have no incentive to make the amount of troops
they have ready to fight public, as turning the game into a sequential one with
the COIN forces moving first would result in a rollback equilibrium, leading to an
outcome (B2 ) smaller than the expected payoff they would receive when the game
is played simultaneously, i.e.
COIN
ESurge D B1  q C B2  .1  q/ > B2 ; because B2 < B1 . (7.10)

While the COIN forces know whether they can play SURGE or not, the
insurgents are not aware of the kind of forces they will meet in case they decide to
fight. Let us now assume, given the pertinent opportunity cost restrictions, playing
SURGE is only possible with a small probability w. The insurgents will decide to
fight or withdraw by comparing their respective expected payoffs
INS
EFight D w  R1 C .1  w/  R3 and
(7.11)
INS
EWithdraw D w  R2 C .1  w/  R4 respectively.

The payoff matrix, given that no SURGE troop levels would be available for the
COIN forces—which, however, would not be known to the insurgents—is given in
Fig. 7.6.
The matrix shows the expected payoffs of the players, who are endowed with
different amounts of information. In the top row, the COIN forces know that the
operation is only a RAID. The insurgents, on the other hand, do not know whether
they will face a RAID or a SURGE, and will therefore take the respective expected
7.4 Surge: An Enemy-Centric Approach 75

Fig. 7.6 Payoff matrix for the enemy-centric surge with asymmetric information

payoffs into account. In the bottom row, the insurgents’ payoff from FIGHT would
be the one from facing a SURGE, because this would be the only kind of operation
ever to be launched. The RAID-only COIN forces, on the other hand, would always
know their resulting respective payoffs for certain: In the top row it would be either
B3 or B4 , while in the bottom row the status quo would prevail throughout.
The kind of resulting Nash equilibrium depends on how the insurgents respond
to the launch of a RAID—though they would not know whether it will turn out to
be a RAID or a SURGE. Comparing their expected payoffs in (7.11), the insurgents
will decide to choose FIGHT only if

R3  R4
w< D: w(7.12) , (7.12)
R3 C R2  R4  R1

i.e. for w being “small” in the sense of (7.12), no Nash equilibrium in pure strategies
exists, implying that there is a Nash equilibrium in mixed strategies; otherwise a
uniquely defined Nash equilibrium could be found in the top right cell.
In the case of (7.12) holding, the COIN forces’ expected payoff would be given
by
   
COIN
EOverall D w  q  B1 C .1  q /  B2 C 1  w  B5 . (7.13)

Now consider the COIN forces switching to a pure strategy in the sense that they
either would never attack with a weak force, i.e. the only kind of operation to be
launched would be SURGE. The COIN forces would then position themselves as
a first mover. In the first case, the insurgents would sooner or later learn, that once
an operation has been launched, it can only be a SURGE and would retreat in time,
yielding to an overall COIN forces’ payoff of
COIN
EOverall D w  B2 C .1  w/  B5 , (7.14)

which, due to (7.10), is smaller than the one given by (7.13).

However, with such a strategy, i.e. only attack if it is reasonably safe to do so, the
COIN forces would never be able to put an end to the insurgency as the insurgents
would always be able to elude what Gerry Long called “decisive” battle (Long
76 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

2014). Therefore, COIN forces could alternatively choose the other pure strategy,
which is to always launch an operation, irrespective of the troop numbers available.
The idea would be to tempt the insurgents into at least not always withdrawing once
they observe an operation going on. In that case, the insurgents would always offer
a fight, due to (7.12), implying they could be lucky if the operation turns out to be
only a RAID—in which case those COIN forces would suffer heavy casualties—, or
they would come close to being annihilated as they would not stand a chance against
a vastly superior enemy. Even though this would mean that the COIN forces would
suffer strong casualties from time to time, following the enemy-centric approach,
only by fighting the insurgents would they fulfill their ultimate objective.
The expected payoff for the COIN forces by always launching an operation while
the insurgents always fight would thus be given by
COIN
EOverall D w  B1 C .1  w/  B3 . (7.15)

First, when comparing the two pure strategies, i.e. always launching an operation,
irrespective of the type, with never launching a RAID, we find (7.15) would be
greater than (7.14) if and only if

B5  B3
w> D: w(7.16) . (7.16)
B1  B2  B3 C B5

In order to compare the overall payoffs of always launching a RAID with that of
the mixed strategy, we first exploit the indifference method. As COIN forces would
be indifferent between playing the mixed strategy and playing any other strategy,
this would hold if we compare the expected conditional payoff when playing a
mixed strategy with the non-random payoff resulting from the pure strategy of never
launching a RAID. As this implies
COIN
ERaid D q  B3 C .1  q/  B4 D ENo Raid D B5 ,
COIN
(7.17)

the mixed strategy equilibrium probability is given by q D BB44 B

B3
5
.

Substituting q into (7.13) and comparing (7.13) with (7.15) we find
!
B  B  B  B 
4 5 5 3
w B1 C B2 C.1w/B5 7 wB1 C.1w/B3 (7.18)
B4  B3 B4  B3

which is equivalent to
 
w B5 .B1 CB2 B4 CB3 /CB3 .B1 B2 CB4 B3 / 7 .B4 B3 /.B3 B5 / (7.19)

or
B4  B3
w7 . (7.20)
B1  B2 C B4  B3
7.4 Surge: An Enemy-Centric Approach 77

Table 7.4 COIN strategies Operation Always Sometimes No RAID

ranking
w < w(7.16) 3rd best 2nd best Best
w(7.16) < w < w(7.21) 2nd best Best 3rd best
w > w(7.21) Best 2nd best 3rd best

Thus, COIN forces would be better off by always launching an operation rather
than just from time to time, as would be the case in the mixed strategy equilibrium,
only if

B4  B3
w> D: w(7.21) . (7.21)
B1  B2 C B4  B3

Finally, when comparing the critical thresholds of w(7.16) and w(7.21) , we find that

B5  B3 B4  B3
7 (7.22)
B1  B2  B3 C B5 B1  B2 C B4  B3

is equivalent to .B5  B3 / 7 .B4  B3 /.

Because of (7.8), .B5  B3 / < .B4  B3 /, implying that w(7.16) < w(7.21) holds
throughout, which provides exact rankings of the three strategies depending on the
value w takes in relation to w(7.16) and w(7.21) : If w is “large”, i.e. w > w(7.21) , the best
COIN strategy would be to always launch whatever operation is available, followed
by the mixed strategy. If w takes some “medium” value, i.e. w(7.16) < w < w(7.21) ,
the COIN forces’ best option would be to play a mixed strategy, i.e. to launch a
RAID sometimes and to abstain at other times, followed by going first and always
launching an operation. Finally, if w is “small”, i.e. w < w(7.16) , the COIN forces
would be best advised to never launch any RAID, followed by sometimes launching
a RAID. Summarizing, we find the rankings to be as shown in Table 7.4.
Irrespectively, though, w should remain small enough, i.e. (7.12) should hold, in
order for the insurgents to attack in the simultaneous game as no resistance would
ever materialize.
At first, it seems difficult to compare w(7.12) with w(7.16) or w(7.21) as they refer to
the preferences of different players. Still, if we take a closer look at the payoffs, it
seems reasonable to assume that the payoff either force receives when “winning”,
i.e. getting its best possible result, should outrank its next possible payoff by some
margin. Accordingly, we would assume that the best outcomes in the payoffs
distribution, i.e. B1 and R4 respectively, should carry, according to their payoff
distribution, a weight significantly higher than the other outcomes. As this would
be equivalent to R3  R4 being larger than B4  B3 and R2  R1 being smaller than
B1  B2 . Under these circumstances we can presume that w(7.12) might well be larger
than w(7.21) , and thus condition (7.12) to hold throughout.
The table above can also be used to make decisions about operations even in
the absence of perfect information about w. There are a number of reasons why a
commander may not have perfect information about this value; these range from
78 7 Modelling Specific Aspects of COIN Operations: On the Possibility of a First. . .

communication issues to the fact that a decision about increasing the amount of
troops can be related more to political than military affairs.
In cases where we do not have any other information about the value of w apart
from knowing that condition (7.12) still holds, the table can be used to help to design
a kind of overall max-min-strategy: It would, interestingly, be a mixed one, even
though it carries the risk of heavy COIN casualties if the occasional RAID meets
significant resistance. In contrast, never launching an operation would be likely to
lead to the worst possible outcome.

7.5 Conclusions

All the aforementioned models aim to explain what kind of strategies would be best
suited for a scenario comparable to the situation in Iraq at the beginning of 2006.
The main difference lies in the ultimate objective of the COIN forces, and therefore
in the COIN forces’ preferences.
There are two important aspects that distinguish these models. First, while in
the first three models the distribution of the US forces’ preferences was determined
by accomplishing an immediate but lasting peace—note that the payoffs resulting
from the insurgents’ withdrawal were preferred over all the others—the last model is
very different, as the final objective is the military defeat of the insurgency14 which
necessarily implies confrontation. Or, to put it another way, the difference in the
payoffs’ distributions depends on what is considered a priority by the strong force:
To achieve an immediate peace, or to put an end to the insurgency, and, secondly,
on the information the different players have at different stages of the game.
The first four models show that the strong force would achieve its best outcomes
by moving first, as the insurgents are prevented from making any strategic move due
to their inability to commit. Moving first means revealing information and being
at ease with any information that could have been obtained later. These models
would support the idea of COIN forces executing their strategy without any further
consideration about the enemy’s moves. Even though, obviously, the final outcome,
when moving first, would still depend on the enemy, COIN forces basically force
the insurgents onto a branch of the game, the COIN forces would prefer.
Another interesting result unfolds when considering the last model, as it is a zero-
sum game without a possible first mover advantage, if viewed from the perspective
of Chap. 4. However, by introducing a stochastic restriction constraining the COIN
forces’ strategies, an asymmetry of information is created, which at the end allows
for the possibility of a first-mover advantage after all.

14
Note that in the model described in Sect. 7.3 any long lasting solution to the conflict would also
be preferred to the rest of the options (i.e. all strategies with CHB) but because of the people-centric
approach, a solution in these terms avoiding confrontation would be even better.
7.5 Conclusions 79

Table 7.5 Summary of findings

Model Payoff preferences Information Best strategy
Section 7.1.2 SURGE Short-term Perfect First mover
Section 7.2.1 CHB Short-term Perfect First mover
Section 7.3 CHB & SURGE Long-term Perfect First mover
Section 7.4.1 Enemy-Centric Long-term Asymmetric Mixed

While in the first models the COIN forces were interested in conveying the truth
about their behavior and thus moving first, the last model epitomizes the importance
of employing a mode of decision that avoids any deductive anticipation by the other
player. Furthermore, following the modeled behavioral pattern, the COIN forces
will not only be interested in anticipating the insurgency strategies, but they will be
conditioning them.
Table 7.5 summarizes the most important features of these models.
Chapter 8
The War Against the Taliban: Tactical
Operations and Strategic Moves

8.1 Introduction

The Afghanistan war exhibits two special features. First, it is exemplified not
only as the most recent example of “Irregular Warfare”, or IW, but as the prime
example for the kind of armed conflict the West has to expect in the twenty-first
century. Whichever definition of IW one looks at, its paramount characteristic is
a (strong) asymmetry of the opponents’ objective functions. In order to achieve
overall, i.e. politically sustainable, victory, it no longer suffices to defeat the enemy
army but other aspects such as culture, religion or the problem of dealing with
less perfect Third World-style “democracies” in general have to be taken into
account. In particular, IW today is about the—almost mantra-like—“winning the
hearts and minds”1 of the civilian population, which becomes all the more difficult
if that civilian population does not feel represented by its, i.e. the host country’s,
government.2 Further, as on the one hand the political support in a parliamentary
democracy for conducting a war in a faraway country dwindles with every body bag
arriving home and, on the other hand, in order to neither unduly damage the position
of the host nation’s government nor offend the host nation’s culture, the footprint left

This chapter appears in the International Journal of Strategic Decision Sciences, authored by Lucía
Martínez Ordóñez and Jörg Schimmelpfennig (Martínez and Schimmelpfennig 2015b), Copyright
2011, IGI Global, www.igi-global.com. Posted by permission of the publisher. A modified version
of the model, referring to counterinsurgency warfare in general, can be found in Martínez and
Schimmelpfennig (2015a).
1
The phrase goes back to the British General Sir Gerald Templer who, as British High Commis-
sioner, was in charge of the British Army counterinsurgency campaign in Malaya from 1952 to
1954. See, e.g., Nagl (2002).
2
The earliest examples of warfare in the history of mankind typically show at least some features
of IW. One of the more prominent examples from recent history would of course be the American
war of Independence. Still, IW as a doctrine did not evolve until the Vietnam War.

© Springer International Publishing AG 2017 81

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_8
82 8 The War Against the Taliban: Tactical Operations and Strategic Moves

by the occupying army not only has to be limited in terms of size, but in terms of the
duration of the campaign, too, time would be on the side of the insurgents as well.3
It is the second feature that gave rise to relying on so-called Forward Operating
Bases, or FOBs.4 Their prime purpose has been two-fold: on the one hand, by their
mere presence right in the middle of Taliban-controlled areas, they were supposed to
reassure the civilian population that ISAF, by not just hiding within heavily fortified
garrisons, was not only present but also willing to take on the Taliban in order to
liberate in due time Afghan villages from the Taliban stranglehold; on the other
hand they were a means to project power into Taliban-held territory by providing
a (more or less) safe haven from which tactical operations could be launched on a
routine basis.5
To simplify matters, ISAF operations are assumed to be of two types only.
Whatever the exact purpose of an operation, in an ideal world it should follow a pre-
planned schedule. It would not suffice to rely on close air support to be delivered
by fixed-wing aircraft alone. Insurgents would be at an advantage once it comes to
“danger close” fighting because fixed-wing aircraft could no longer deliver close
air support without endangering the lives of the very troops they are supposed to
protect. They would thus have every incentive to attack at close quarters thus making
ISAF forces’ casualties more likely. Attack helicopters should therefore be assigned
in advance and made to wait on station. In reality, though, as, e.g. the experience
from the Afghanistan war has shown, such helicopters may not always be available
in sufficient numbers, and quite often they could only take to the air once troops
in contact has been declared. The resulting delay implies that the troops on the
ground would be without air support when it is mostly needed. An ISAF operation
would thus either have attack helicopters assigned in advance, or, due to the lack of
resources, it would not.
Once an operation has been launched, the Taliban would have to decide whether
to hold the ground and to ambush-style fight ISAF forces, or to withdraw. At the
moment they have to make up their mind, though, they would not know whether

3
Mujahid Rahman, a Taliban fighter, once said: “Americans have the watches; we have the time”,
thus indicating the need for ISAF to achieve their objectives subject to a time limit.
4
The strategy looks similar to the one employed by the Romans some two millennia ago where
“castra” were one of mainstays when, what they called it, “pacifying” other European countries.
Roman strategy, though, went much further than ISAF’s in Afghanistan: while the latter try to
leave a as small as possible footprint in their host country, Rome never tried to “win the hearts and
minds” of the indigenous population: if opposed, they simply killed all fighting-age males and sold
women and children as slaves; if not, they tried to assimilate the indigenous population by offering
Roman citizenship on a large scale while at the same time leaving local customs and religions as
they were. Still, Roman warfare would thus classify as one the earliest examples of IW, too, as
basically Roman troops usually did not have to fight regular enemy armies but far more often had
to deal with local insurgencies.
5
Whether or not surrendering the one distinct advantage ISAF has over the Taliban, which is its
ability to mobility in general and mobile armour in particular and thus conduct a much more
mobile “Manoeuvre Warfare”, is open to debate. For a more comprehensive discussion of the FOB
strategy, see Long (2014) and Owen (2014).
8.2 Scenarios and Evaluation 83

attack helicopters are waiting just over the horizon. The ISAF commander would
of course know about the availability of attack helicopters. If they are available, he
would launch the operation because this is what the scenario is all about. If they are
not, he would have to decide whether to go on anyway or to refrain from launching
the operation. He would not know, though, how the Taliban are going to react.
The situation is very much in line with the Clausewitzian conjecture that whoever
knows about the decision taken by his opponent, or the options he has to be content
with, would be expected to enjoy some kind of advantage. The Taliban would be
happy to learn about the kind of air support available to ISAF in advance, but all
they can do is to make an educated guess from their past experiences. Likewise the
ISAF commander would be eager to know the Taliban tactics of the day.
The purpose of this model is to analyze the situation by modelling it as a
game with asymmetric information.6 In particular, due to the “Irregular Warfare”
nature of the situation, the availability of strategic moves is examined in order to
establish whether such a situation would still be compatible with the conventional
Clausewitzian paradigm, or if either side could enjoy some kind of first-mover
advantage.7

8.2 Scenarios and Evaluation

ISAF operations can be of two types: either sufficient close air support has been
assigned in advance (CAS), or, due to the lack of resources, the operation has to
be conducted without close air support (w/o CAS). On the other hand, the Taliban
can choose between either trying to fight back and hold the ground (FIGHT) or
withdrawing (W/DRAW). Hence, four possible scenarios, denoted clockwise by
Roman numerals, can be identified (Fig. 8.1).

6
Traditionally, battlefield situations have been viewed as zero-sum games, i.e. the sum of the (two)
opponents’ payoffs is equal to zero for every outcome of the game. For all practical purposes, zero-
sum games are equivalent to constant-sum games—the sum of the payoffs would be identical to
some predetermined number such as, if the respective outcomes are measured as the probability
to win—or even games which can be transformed into zero-sum games by applying the same
strictly monotone transformation to every payoff. Zero-sum games implicitly form the basis of the
“Network Centric Warfare”, or NCW, doctrine prevalent in today’s military thinking presumably
giving rise to the Clausewitz metaphor of “Der Nebel des Krieges”, or the “fog of war”, leading to
a second-mover advantage, i.e. being able to learn about any move made by the opponents would
be preferable over having to move first, i.e. without having access to such respective information;
see, e.g. Friedman (2009).
7
In order not to confuse any reader who is more accustomed to the military understanding of
“strategy”, in contrast to “tactics”, it should be stressed that the wording used here is the game
theoretic one. Any move, even though representing different military tactics, is called a strategy,
and any move actually changing the nature of the game, even though it would still involve changing
military tactics, would be called a “strategic move”.
84 8 The War Against the Taliban: Tactical Operations and Strategic Moves

Fig. 8.1 Payoff matrix for Taliban

ISAF operations
FIGHT W/DRAW

CAS I II

ISAF

w/o CAS III II

In addition, there is of course an outside option IV, describing the case where
ISAF does not launch any kind of operation. In order to compare the respective
payoffs, the four scenarios are looked into in more detail:
I (CAS + FIGHT) Because attack helicopters tip the scales, ISAF forces will
achieve a tactical victory. However, there are casualties on both sides.
II (CAS + W/DRAW) As the Taliban withdraw, the objective of the mission is
achieved irrespective of the availability of attack helicopters because no “danger
close” situation could occur. Casualties are low or even non-existent, resulting
in a resounding tactical victory by ISAF.
III (w/o CAS + FIGHT) Without attack helicopters being on the scene there would
be no chance to provide air support in time for “danger close” situations, ISAF
forces suffer heavy casualties and eventually have to retreat. The Taliban hold
the ground, resulting in a resounding Taliban tactical victory.
IV (no operation) The status quo prevails.
Denoting ISAF payoffs by ˆ, with the subscripts referring to the four scenarios,
these can be ranked as

ˆII > ˆI > ˆIV > ˆIII 8 . (8.1)

Denoting the Taliban payoffs respectively by ,

III > IV > II > I .9 (8.2)

8
In particular with respect to the first of these inequalities, one should remember that ISAF
refrained from using the traditional war metric of counting enemy casualties—which so disas-
trously contributed to the misapprehension of the military situation during the Vietnam War—and,
instead of trying to kill as many Taliban as possible, just attempted to achieve its respective military
objective with the minimum number of ISAF casualties.
9
While killing enemy fighters in general and ISAF troops in particular—one must not forget that, as
the British painfully learned during their nineteenth century Afghanistan campaign, Afghans have
always been anything but reluctant when it comes to fighting invaders, or, as a BBC defense analyst
once put it: “Afghans will happily fight on for another 50 or 100 years. It’s their favourite pastime
anyway”—has certainly stood high on the Taliban agenda, this does not imply that they Jihadis
eager to sacrifice their own lives for their cause no matter what. On the contrary, a significant
number of Taliban fighters, colloquially called “$10 Taliban” by ISAF soldiers, are only fighting
8.3 Solving the Simultaneous Game 85

Due to the IW nature of the conflict, no restriction has been imposed on the
respective sums of the ISAF and Taliban payoffs. Moreover, due to ˆII > ˆI and
II > I , the game could not be zero-sum anyway.

8.3 Solving the Simultaneous Game

Assuming that neither ISAF would know the Taliban strategy at the time they
had to decide on whether to undertake an operation nor would the Taliban know
whether rotary-wing close air support will be available to ISAF forces, the game is
a simultaneous one. Obviously, due to (8.1), if close air support is available, ISAF
would always opt for conducting the operation as either payoff, i.e. ˆI or ˆII , is
higher than ˆIV .
Now let w be the probability that rotary-wing close air support is available. Then,
in order to decide on whether to choose FIGHT or W/DRAW, the Taliban would
have to compare the respective expected payoffs, i.e.

ETaliban .FIGHT/ D w  I C .1  w/  III , resp.

(8.3)
ETaliban .W=DRAW/ D II .

As the former would be smaller than the latter if and only if

III  II
w> , (8.4)
III  I

the Taliban, without any additional information, would, once an operation has been
launched, always withdraw, implying that ISAF would always mount an operation
irrespective of whether close air support is available or not. Now turn to the case of
w being sufficiently small, i.e.

III  II
w< , (8.5)
III  I

which would of course imply

ETaliban .FIGHT/ D w  I C .1  w/  III > E.W=DRAW/ D II . (8.6)

The resulting payoffs are given in Fig. 8.2.

It is important to visualize that because of the asymmetric information, the
scenarios in each of the two left-hand cells relate to two different perspectives,

part-time: basically they are just peasants eager to earn some money following the harvesting but
return to and look after their families once the new farming season begins. See, e.g., Tootal (2009).
86 8 The War Against the Taliban: Tactical Operations and Strategic Moves

Fig. 8.2 Payoff matrix for ISAF operations with asymmetric information

implying the two sides have to expect different scenarios in two of the four cells.
In the top left cell, the ISAF commander, knowing that the operation would have
be launched without attack helicopters, would, assuming that the Taliban decide to
hold, know for certain that the outcome will be that of scenario III. The insurgents
on the other hand are unaware of the composition of the ISAF forces and would thus
not be able to predict whether the outcome will be scenario I or scenario III. In the
bottom left cell, without any operation, the status quo, i.e. scenario IV, obviously
prevails, irrespective of whether the Taliban would have tried to hold the ground.
However, as the Taliban can only commit to hold the ground once an operation has
actually been launched, that operation would command proper air support—in the
bottom row these are the only type of operations going on—and the Taliban would
thus be brought face to face with attack helicopters resulting in scenario I (Fig. 8.2).
In order to learn about an equilibrium, every cell is compared with its respective
neighbour. Assume first that ISAF forces would never launch any kind of operation
unless attack helicopters have been assigned. Then the only kind of operation ever
to take place would involve attack helicopters. As they rate II over I, the Taliban,
learning from their experiences, would withdraw. Then, however, ISAF forces could
well launch an operation even without attack helicopters, since they would not
meet any resistance: they prefer II over IV. In turn, though, the Taliban would soon
learn that no attack helicopters are coming and would, as they rate the average of
scenarios I and III higher than scenario II, decide to hold the ground no matter what.
Finally, in case the insurgents always provide resistance, ISAF forces would be well
advised to step back from launching an operation unless “danger close” air support
can be guaranteed because they rate IV over III—which is the very same scenario
the analysis began with.
Summarizing, no saddle point exists, and thus no Nash equilibrium in pure
strategies either.10 Then, as the number of Nash equilibria has to be odd, a mixed-
strategy Nash equilibrium has to exist: ISAF would, provided no attack helicopters

10
While game theorists prefer the concept of the Nash equilibrium, respective military textbooks
use the concept of saddle points; see, e.g. Wagner et al. (1999). However, the two concepts are
equivalent in the sense that whenever there is only one Nash equilibrium, it would only be in pure
strategies if it is a saddle point, too.
8.4 Reality Check 87

are available, mount an operation with some probability p—remember that with
CAS ISAF always goes for OPERATION—and with a probability of 1  p it would
not, while the Taliban would choose FIGHT with some probability q and W/DRAW
with a probability of 1  q. The purpose of such “mixed strategies” obviously is
to avoid putting oneself at risk of being outwitted by the enemy. Hence, whenever
either side embarks on such a mixed strategy, the other side would not be able to
gain any advantage from changing its strategy and, by argumentum e contrario,
would thus be indifferent between all of its options, including playing the mixed
strategy.
Such an equilibrium could of course be easily computed by, e.g., using the indif-
ference method (see, e.g., Dixit and Skeath 1999). The result, i.e. the equilibrium
values p and q would of course depend on i , i D I; II; III, and ˆj , j D II; III; IV,
respectively. Irrespective of the argument that, as the respective enemy’s payoffs
would not be known to either side, only a more or less reasonable estimate would
have to do, it is not necessary to know the results for the purposes of this model as
the resulting payoffs would still be known due to the indifference method argument:
whatever values p and q take, the resulting expected payoffs for either player
would be identical equal to those resulting from playing any other strategy, including
his two pure strategies, provided the other player sticks to his equilibrium mixed
strategy. Thus, ISAF’s expected payoff would be given by

EISAF D ˆIV , (8.7)

i.e., the one from playing its pure strategy “No Operation”, while the Taliban payoff
would be given by

ETaliban D II , (8.8)

i.e. the one resulting from always playing W/DRAW whenever an operation has
been mounted.

8.4 Reality Check

The results from the model suggest that ISAF would still have mounted operations
from time to time even without rotary-wing close air support having been assigned
in advance, while the Taliban would sometimes have tried to hold the ground,
and sometimes to withdraw in time. It is compatible to what actually happened in
Afghanistan. While statistics on such incidents are classified and thus unavailable in
the public domain,11 personal recollections of ISAF veterans suggest that “danger

11
Note that the Wikileaks Afghan War Diaries (Wikileaks 2010) are of no use either due to the
poor categorization and arbitrariness of the sample.
88 8 The War Against the Taliban: Tactical Operations and Strategic Moves

close” situations were anything but a mere exception (see, e.g., Duncan 2011;
Lewis 2010 or Tootal 2009). On an almost regular basis, due to a shortage of
attack helicopters, patrols were sent into Taliban-controlled areas without sufficient
close air support. Sometimes they met fierce resistance, and sometimes the Taliban
were long gone. On the other hand, even when proper close air support was
available from the very moment an operation had been launched, the Taliban did
occasionally attempt to hold the ground, though futilely, and subsequently suffered
heavy casualties.

8.5 A Wider View: Strategic Moves?

There are just two ways to make a strategic move: either a move is introduced or
the game is changed from a simultaneous one to sequential one. The first idea is
obvious: as w is, from the perspective of the Taliban, but a subjective probability,
i.e. a more or less educated guess, ISAF could try to overstate its abilities in order to
make the Taliban believe that proper CAS would be available on the day. Respective
instruments would be never to hide, underestimate or downplay its CAS capabilities
and/or availability, and/or try to deter by doing a “show of force”, i.e. low-altitude
high-speed passes of more readily available fixed-wing aircraft, thus suggesting a
much stronger CAS force which hopefully would, provided the critical point given
by (8.4) could be reached, result in a Taliban withdrawal. The model thus provides
a game-theoretic rationale for deception, and, in particular, never to downplay one’s
own strength.12 However, it was easier said than done; as personal experiences from
the ground tell, some of the time the Taliban did indeed fall for it, but over time they
had learned that a fixed-wing platform is anything but a substitute for a rotary-wing
platform and ISAF did not attempt a show of force for nothing: it was all they had
(see, e.g., Lewis 2010).
The other option, turning the game into a sequential one, would require the
respective player to become the first mover thus effectively assigning the role of
the second mover to his opponent. It could of course be done by switching to a
pure strategy: as situations like the ones discussed here are anything but one-offs
but happened on a regular if not daily basis, the opponent would rather sooner than
later learn about the change in tactics and adapt accordingly. It looks anything but
promising: at best, nothing would change; at worst, it runs contrary to the intuitive
idea of the Fog of War originally suggested by Clausewitz and subsequently adopted
by the Network-Centric-Warfare doctrine (see, e.g., Friedman 2009) as, without any
kind of uncertainty left, it could be exploited by the enemy just the same any other
strategy than the mixed-strategy equilibrium one can be outwitted. A closer look,
though, reveals otherwise.

12
For a much wider discussion of the benefits of military deception, see Elkus (2012).
8.6 Concluding Remarks 89

Assume, that ISAF decided to put a halt to all operations that cannot be
supported by attack helicopters. The Taliban would update their beliefs accordingly.
In particular, they should learn, as argued above, that from now on every single
ISAF operation would always be accompanied by attack helicopters and would
wisely back off from offering any resistance. Previously ISAF’s expected payoff13
had been equal to that of scenario IV because all strategies would yield respective
identical expected payoffs to either player in any mixed-strategy equilibrium. The
expected payoff from the top line in the matrix would thus have been identical to
the guaranteed payoff from the bottom line. The new first-mover strategy would
directly lead to the payoff from the bottom line as given by (8.7). Turning to the
case of attack helicopters being available, though, there would be an improvement.
Previously every ISAF operation, whether supported or not, had come under attack
from time to time—it was an equilibrium in mixed strategies—as the Taliban could
not tell the difference in advance, resulting in an overall ISAF payoff of

EISAF D w  .ˆI  q C ˆII  .1  q // C .1  w/  ˆIV . (8.9)

With the first-mover strategy, the only operations going on would be supported
by attack helicopters. They would no longer meet any resistance because the Taliban
would by now have decided never to offer resistance anymore. ISAF’s payoff thus
becomes

EISAF D w  ˆII C .1  w/  ˆIV . (8.10)

Due to ˆII > ˆI , any linear combination of the two would be smaller than ˆII ,
implying that ISAF’s payoff has increased, thus handing a first-mover advantage to
ISAF when choosing never to mount an operation without proper CAS. Not only
would the tactical objective be accomplished, but there would not be any casualties
either.

8.6 Concluding Remarks

It has long been argued elsewhere (see, e.g., Schwartz 2011) that unless counterin-
surgency operations (COIN) can be guaranteed proper air support, they should not
try to embark on tactical operations. A game-theoretic analysis lends support to
this thesis by showing that doing otherwise would only tempt insurgents to offer
resistance in situations where they were better advised not to. The resulting net gain
from improvised operations, i.e. without proper air support, would, when compared

13
Remember that the Taliban payoffs do not matter to ISAF due to the asymmetry of the objective
functions. Rather, ISAF’s only concern should be whether it would be better off by whatever new
equilibrium emerges. In case no attack helicopters are available, nothing would actually change.
90 8 The War Against the Taliban: Tactical Operations and Strategic Moves

to not running any such operation, on average be zero. However, properly supported
operations would inadvertently be compromised. Insurgent casualties would go
up—one has to remember, though, that in a counterinsurgency situation this is not a
primary objective anyway—but so would COIN forces casualties, and this is what
should be of concern to both COIN commanders as well as their political masters
alike.
Returning to the “winning the hearts and minds” rationale, a reduced number
of tactical operations could of course be considered unacceptable from an overall
political perspective as it might, e.g., be interpreted as putting COIN forces’ resolve
into question. Then, however, the only alternative is to increase the number of attack
helicopters and crew. Putting COIN forces into harm’s way without such support
should never be an option, and does not make any military sense either.
Chapter 9
Lessons Learned

Military doctrine assumes information and information technology to be a key factor

in decisions at all levels of war. It suggests that one should expect military operations
to go hand in hand with second mover advantages. The analyses presented in this
thesis suggest, though, that the general assumption that every strategy anticipating
the enemy’s intentions would necessarily result in an advantage is not justified. What
is more, we even find examples in actual conflicts, where there is an incentive to do
the opposite, namely reveal one’s strategies to the enemy. Scenarios from Operation
Enduring Freedom, e.g., show that, contrary to intuition, COIN forces’ operations
would have been able to do better by revealing their CAS capabilities in advance of
every operation to take place, and not going ahead with that particular operation in
the first place, unless they possess sufficient capabilities.
Turning to Operation Iraqi Freedom, we find similar results. There are two
features that conditioned COIN operations in Iraq when the spiral of violence
peaked in 2006: “The Surge” and changing US COIN doctrine from an enemy-
centric to a population-centric approach. A game theoretical analysis shows that the
strategies leading to the best possible outcomes for the COIN forces are independent
of the expectations they might have about the strategies of Al Qaeda in Iraq. Thus,
whether to mix their strategies or move ahead would depend exclusively on the final
objectives of the COIN forces, i.e. provide security or put an end to the insurgency.
While there is no doubt that information technology provides valuable tools
to both military planners as well as the warfighter on the ground, strategic
planning could, at least in some cases, benefit from a look beyond the traditional
Clausewitzian horizon. The degree to which this insight has already entered, or
might enter, the military planner’s mindset is difficult to assess. It seems as if there
is a long way to go, though, as the most recent reference available to this work
illustrates: During a recent conference on the Vietnam War, one author summarized
his findings—from an analysis of the 1972 bombing campaigns against North
Vietnam that aimed to convince North Vietnam to return to the negotiating table

© Springer International Publishing AG 2017 91

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0_9
92 9 Lessons Learned

on America’s terms—as the understanding of “strategic planning as a competitive

iterative process” (Randolph 2015).
It is an indubitable fact that strategies—and in the game-theoretic sense this
means just actions—may well lead to one’s opponent modifying his strategies. It
is a far cry, though, from comprehending the concept of an equilibrium or that of
a second-mover, or, for that purpose, a first-mover advantage, let alone a strategic
move.
Acknowledgements

I want to express my gratitude to my supervisor, Jörg Schimmelpfennig, for his

overall support, for the inspiring discussions of research ideas, for reviewing all my
drafts and for co-authoring part of my work, and especially for making me laugh
when I needed it.
I am also very grateful to Wilhelm Löwenstein for motivating me to finish this
thesis, in a time, when throwing in the towel appeared to be an attractive possibility.
I deeply thank my parents, Avelino y Adela, and my sister Paula, for their
unconditional trust, encouragement and love.
Finally, I thank Götz Freytag, who is there for me when it counts.

© Springer International Publishing AG 2017 93

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0
References

Air-Sea Battle Office (2013). Air-sea battle. Service collaboration to address anti-
access & area denial challenges. http://www.defense.gov/Portals/1/Documents/pubs/ASB-
ConceptImplementation-Summary-May-2013.pdf. Last accessed 18 Oct 2015.
Alexander, J. H. (2012). Close air support: The pioneering years. http://www.usni.org/magazines/
navalhistory/2012-11/close-air-support-pioneering-years. Last accessed 10 Oct 2015.
Archick, K., Belkin, P., Blanchard, C. M., Humud, C. E., & Mix, D. E. (2015). European fighters
in Syria and Iraq: Assessments, responses, and issues for the United States. Retrieved from
Congressional Research Service: https://fas.org/sgp/crs/row/R44003.pdf. Last accessed 18 Oct
2015.
Armstrong, B. (2015). D-all the above: Connecting 21st century naval doctrine to strategy. Infinity
Journal, 4(4), 13–17.
Arreguín-Toft, I. (2001). How the weak win wars. A theory of asymmetric conflict. International
Security, 26(1), 93–128.
Aumann, R. J. (1976). Agreeing to disagree. The Annals of Statistics, 4(6), 1236–1239.
Bartholomees, J. B. (2010). The issue of attrition. Parameters, 40(1), 5–19.
BBC news (2012). Syria conflict: Damascus suburb sees heavy fighting. http://www.bbc.co.uk/
news/world-middle-east-19645269. Last accessed 18 Oct 2015.
Belasco, A. (2014). The cost of Iraq, Afghanistan, and other global war on terror operations since
9/11. Retrieved from Congressional Research Service: https://fas.org/sgp/crs/natsec/RL33110.
pdf. Last accessed 18 Oct 2015.
Belasco, A. (2015). Defense spending and the budget control act limits. Retrieved from Congres-
sional Research Service: https://fas.org/sgp/crs/natsec/R44039.pdf. Last accessed 18 Oct 2015.
Bernheim, B. (1984). Rationalizable strategic behaviour. Econometrica, 52(4), 1007–1028.
Bernheim, B., Peleg, B., & Whinston, M. (1987). Coalition-proof Nash equilibria i: Concepts.
Journal of Economic Theory, 42(1), 1–12.
von Bertalanffy, L. (1968). General system theory. New York: George Braziller.
Betts, R. K. (2002). The soft underbelly of american primacy: Tactical advantages of terror.
Political Science Quarterly, 117, 19–36.
Beyerchen, A. (1992). Clausewitz, nonlinearity, and the unpredictability of war. International
Security, 13(3), 59–90.
Biddle, S., Friedman, J. A., & Shaphiro, J. N. (2012). Testing the surge. why did violence decline
in Iraq in 2007? International Security, 37(1), 7–40.
Bilmes, L., & Stigliz, J. E. (2011) The long term costs of conflict: The case of the iraq war in the
great recession. In D. L. Braddon, & K. Hartley (Eds.), Handbook on the economics of war (pp.
293–307). Cheltenham: Edwad Elgar.

© Springer International Publishing AG 2017 95

L. Martínez Ordóñez, Military Operational Planning and Strategic Moves,
Contributions to Economics, DOI 10.1007/978-3-319-56108-0
96 References

Bilmes, L. J., & Stigliz, J. E. (2008). The three trillion dollar war: The true costs of the Iraq war.
New York: W.W. Norton & Company.
Blank, S. J. (2003). Rethinking asymmetric threats. Carlisle: Strategic Studies Institute of the U.S.
Army War College.
Boulding, K. E. (1962). Conflict and defense a general theory. New York: Harper & Row.
Bradley, O. N. (1951). A Soldier’s story. New York: Henry Holt and Company, Inc.
Brauer, J., & Tuyll, H. V. (2008). Castles, battles and bombs: How economics explains military
history. Chicago: The University Chicago Press.
Bueno de Mesquita, B. (1981). The war trap. New Haven: Yale University Press.
Bueno de Mesquita, B. B., & Lalman, D. (1992). War and reason. New Haveb: Yale University
Press.
Butler, C., & Gates, S. (2010). The technology of terror: Accounting for the strategic use of
terrorism. Economics of Security, Working Paper 30, pp. 10–24.
Cebrowski, A., & Garstka, J. (1998). Network-centric warfare: Its origins and future. Proceedings
Magazine of the U.S. Naval Institute, 124/1/1(139), 28–35.
Chadefaux, T. (2011). Bargaining over power: When do shifts in power lead to war? International
Theory, 3(2), 228–253.
von Clausewitz, C. (1976). On war. Princeton: Princeton University Press. Transls. M. Howard, P.
Paret.
Cochrane, M. (2009). Iraq report 12: The fragmentation of the sadrist movement. Washington:
Institute for the Study of War.
Cockburn, P. (2008). Who is whose enemy? London Review of Books, 30(5), 14.
Congressional Budget Office (2008). Contractors’ support of U.S. operations in Iraq. Last accessed
08 Oct 2015.
Daase, C. (2007). Clausewitz and small wars. In H. Strachan, & A. Herbert-Rothe (Eds.),
Clausewitz in the twenty-first century (pp. 182–195). Oxford: Oxford University Press.
Dahl, E. J. (2002). Network centric warfare and the death of operational art. Defence Studies, 2(1),
1–24.
Department of Defense (2001). Network centric warfare. http://www.dodccrp.org/files/ncw_report/
report/ncw_main.pdf. Last accessed 18 Oct 2015.
Department of Defense (2007). Irregular warfare (iw) joint operating concept (joc), version
1.0. http://www.dtic.mil/doctrine/concepts/joint_concepts/joc_iw_v1.pdf. Last accessed 16 Oct
2015.
Department of Doctrine and Training Development (1976). Mine/countermine operations at
the company level, fm 20–32. http://www.marines.mil/Portals/59/Publications/FM%2020-32
%20W%20CH%201-4.pdf. Last accessed 18 Oct 2015.
Department of Peace and Conflict Research (2015). Website: Definitions. http://www.pcr.uu.se/
research/ucdp/definitions/. Last accessed 18 Oct 2015.
Department of the Army (1993). Field manual 100-5. Operations. http://www.fs.fed.us/fire/
doctrine/genesis_and_evolution/source_materials/FM-100-5_operations.pdf. Last accessed 09
Oct 2015.
Department of the Army (2006). Field manual 3-24 Counterinsurgency. http://usacac.army.mil/
cac2/Repository/Materials/COIN-FM3-24.pdf. Last accessed 18 Oct 2015.
Department of the Army (2008). Fm 3-0. Operations. https://fas.org/irp/doddir/army/fm3-0.pdf.
Last accessed 18 Oct 2015.
Department of the Army (2014). Fm 3-24 Counterinsurgency. On the internet, https://fas.org/irp/
doddir/army/fm3-24.pdf. Last accessed 18 Oct 2015.
Director, A., & Levi, E. H. (1956). Law and the future: Trade regulation. Northwestern University
Law Review, 51(2), 281–296.
Dixit, A. (1979). A model of duopoly suggesting a theory of entry barriers, 10, 20–32. Bell Journal
of Economics, 10(1), 20–32.
Dixit, A., & Skeath, S. (1999). Games of strategy. New York: W. W. Norton.
References 97

Doane, L. M. (2015). It’s just tactics: Why the operational level of war is an unhelpful fiction
and impedes the operational art. Small Wars Journal. http://smallwarsjournal.com/jrnl/art/it
%E2%80%99s-just-tactics-why-the-operational-level-of-war-is-an-unhelpful-fiction-and-
impedes-the. Last accessed 18 Oct 2015.
Dorr, R. F. (2012). Mission to Tokyo. Minneapolis: Zenith Press.
Douhet, G. (1998). The command of the air. Washington D.C.: New Imprint by Air Force History
and Museums Program.
Downey, R., Grubbs, L., Malloy, B., & Wonson, C. (2008). How should the U.S. execute a surge
in Afghanistan? Small Wars Journal. http://smallwarsjournal.com/jrnl/art/how-should-the-us-
execute-a-surge-in-afghanistan. Last accessed 18 Oct 2015.
Duncan, A. (2011). Sweating the metal. London: Hodder & Stoughton.
Elkus, A. (2012). The continuing relevance of military denial and deception. Journal of Military
Operations, 1(1), 21–24.
European Defence Agency (2013). Defence data of eda in 2013. https://www.eda.europa.eu/info-
hub/defence-data-portal/EDA/year/2013. Last accessed 18 Oct 2015.
European Defence Agency (2015). Defence data 2013. https://www.eda.europa.eu/info-hub/press-
centre/latest-news/2015/03/31/latest-defence-data-published. Last accessed 18 Oct 2015.
Fearon, J. D. (1995). Rationalist explanations for war. International Organization, 49(3), 379–414.
Fehr, E., & Schmidt, K. M. (2000). Theories of fairness and reciprocity. In Munich economics
discussion papers (Vol. 2001, pp. 1–46). Munich: Department of Economics University of
Munich.
Fischer, H. (2015). A guide to U.S. military casualty statistics: Operation freedom’s sentinel,
operation inherent resolve, operation new dawn, operation iraqi freedom, and operation
enduring freedom (rs22452; Aug. 7, 2015). https://www.fas.org/sgp/crs/natsec/RS22452.pdf.
Last accessed 18 Oct 2015.
Fravel, M., & Twomey, C. P. (2015). Projecting strategy: The myth of Chinese counter-intervention.
The Washington Quarterly, 37(4), 171–187.
Friedman, B. (2014). Creeping death, Clausewitz and comprehensive counterinsurgency. Military
Review, 94(1), 82–89.
Friedman, N. (2009). Network-centric warfare: How navies learned to fight smarter through three
world wars. Annapolis: Naval Institute Press.
Fudenberg, D., & Levine, D. K. (1993). Self-confirming equilibrium. Econometrica, 61(3), 523–
546.
Gady, F.-S. (2015). Confirmed: China’s defense budget will rise 10.1% in 2015. http://thediplomat.
com/2015/03/confirmed-chinas-defense-budget-will-rise-10-1-in-2015/. Last accessed 18 Oct
2015.
Gentile, G. P. (2009). A strategy of tactics: Population-centric coin and the army. Parame-
ters, 39(3), 5–17.
Gilbert, F. (1986). Machiavelli: The renaissance of the art of war. In P. Paret (Ed.), Makers of
modern strategy: From Machiavelli to the nuclear age (pp. 11–31). Oxford: Oxford University
Press.
Gilboa, I., & Schmeider, D. (1995). Case-based decision theory. The Quarterly Journal of
Economics, 110(3), 605–639.
Grattan, R. F. (2009). The origins of air war: Development of military air strategy in world war I.
London: Tauris Academic Studies.
Hack, K. (2009). The malayan emergency as counter-insurgency paradigm, The Journal of
Strategic Studies, 32(3), 383–Ř414.
Hagopian, A., Flaxman, A. D., Takaro, T. K., Al Shatari, E., & Rajaratnam, S. A. (2013). Mortality
in Iraq associated with the 2003–2011 war and occupation: Findings from a national cluster
sample survey by the university collaborative Iraq mortality study. PLoS Medicine, 10(10),
1–15.
Hall, M. T., & McChrystal, S. A. (2009). http://www.nato.int/isaf/docu/official_texts/
counterinsurgency_guidance.pdf. Last accessed 18 Oct 2015.
Harmon, C. C. (2010). How terrorist groups end. Strategic Studies Quarterly, 4(3), 43–84.
98 References

Hart, B. H. L. (1991). Strategy. New York: Meridian.

Haywood, Jr. O. G. (1954). Military decision and game theory. Journal of the Operations Research
Society of America, 2(4), 365–385.
Herodotus (1926). The persian wars (Vol. IV. Books 8–9). Cambridge: Harvard University Press.
Transl. A. D. Godley.
Heuser, B. (2010). The strategy makers: Thoughts on war and society from Machiavelli to
Clausewitz. Santa Barbara: Praeger/ABC-Clio.
Hoffman, E., McCabe, K., Shachat, K., & Smith, V. (1994). Preferences, property rights, and
anonymity in bargaining games. Games and Economic Behavior, 7(3), 346–380.
Hoffman, F. G. (2007). Conflict in the 21st century: The rise of hybrid wars. Arlington: Potomac
Institute for Policy Studies.
Hoffman, M. (2015). Pentagon bids adieu to air-sea battle name. http://www.dodbuzz.com/2015/
01/21/pentagon-bids-adieu-to-air-sea-battle-name/. Last accessed 18 Oct 2015.
Hume, D. (1985). A treatise of human nature. Book I, Part III. London: Penguin Classics.
Hurley, W., Bucher, C. R., Numrich, S., Ouellette, S. M., & Resnick, J. B. (2009). Non-kinetic
capabilities for irregular warfare: Four case studies, P-4436. Institute for Defense Analysis.
IDA Paper P-4436.
Hurley, W. J., Resnick, J. B., & Wahlman, A. (2007). Improving capabilities for irregular warfare.
Framework and applications (Vol. I). Joint Advanced Warfighting Program. IDA Paper P-4267.
Institute for Defense Analysis.
Institute for Security Studies (2014). Euiss yearbook of European security 2014. http://www.iss.
europa.eu/uploads/media/YES2014.pdf. Last accessed 18 Oct 2014.
Jeffrey, F. J. (2015). Why counterinsurgency doesn’t work. The problem is the strategy not the
execution. Foreign Affairs, 94(Issue 2), 178–180.
Jensen, M. B. (2015). Small forces and crisis management. Parameters, 45(1), 113–124.
Jervis, R. (1978). Cooperation under the security dilemma. World Politics, 30(Issue 2), 167–214.
Joint Chiefs of Staff (2000). Joint vision 2020 America’s military - preparing for tomorrow. http://
www.offiziere.ch/wp-content/uploads/1225.pdf. Last accessed 18 Oct 2015.
Joint Chiefs of Staff (2011). Joint publication 3-0. Joint operations. http://www.dict.mil/doctrine/
new_pubs/jp3_0.pdf. Last accessed 18 Oct 2015.
Joint Chiefs of Staff (2013). Joint publication 3-24. Counterinsurgency. http://www.dtic.mil/
doctrine/new_pubs/jp3_24.pdf. Last accessed 18 Oct 2015.
Joint Chiefs of Staff (2015). Joint publication 1-02. Department of defense. Dictionary of military
and associated term. http://www.dtic.mil/doctrine/new_pubs/jp1_02.pdf. Last accessed 18 Oct
2015.
de Jomini, B. A. H. (1862). The art of war. Philadelphia: J. B. Lippincott & CO. Transl. Capt. G.H.
Mendell, Lieut. W.P. Craighill.
Kagan, K. (2009). The surge, a military history. New York: Encounter Books.
Kaldor, M. (2007). New and old wars: Organized Violence in a Global Era. Stanford: Stanford
University Press.
Katzenbach, E. L., J. (1962). Time, space and will: The politico-military views of Mao-tse-tung. In
T. Greene (Ed.), The guerrilla and how to fight him (Selections from the marine corps gazette)
(pp. 11–21). New York: Frederick A. Praeger.
Keegan, J. (1993). A history of warfare. London: Hutchinson.
Kennedy, J. F. (1962). Remarks at west point (June 6, 1962). Speech, transcript at http://
millercenter.org/president/kennedy/speeches/speech-5766.
Kiras, J. D. (2007). Irregular warfare: Terrorism and insurgency. In J. Baylis, J. Wirtz, C. S. Gray, &
E. Cohen (Eds.), Strategy in the contemporary world (pp. 163–191). Oxford: Oxford University
Press.
Kissinger, H. A. (1969). The Vietnam negotiations. Foreign Affairs, 47(2), 211–234.
Krepinevich, A. F. J. (2015). How to deter China—the case for archipelagic defense.
Foreign Affairs, 94(2). https://www.foreignaffairs.com/articles/china/2015-02-16/how-deter-
china. Last accessed 18 Oct 2015.
References 99

Krulak, C. (2000). Nes cras: Not like yesterday. In R. H. Shultz, & R. L. Pfaltzgraff (Eds.), The
role of naval forces in the 21st century operations (pp. xi–xvii). Washington, D.C.: Brassesy’s.
Kydd, A. H. (2010). Rationalist approach to conflict prevention and resolution. Annual Review of
Political Science, 12, 101–121.
LaGrone, S. (2015). Pentagon drops air sea battle name, concept lives on. http://news.usni.org/
2015/01/20/pentagon-drops-air-sea-battle-name-concept-lives. Last accessed 18 Oct 2015.
Larson, E. V., Eaton, D., Nichiporuk, B., T. S. Szayna, & R. Corporation (2008). Assessing
irregular warfare. a framework for intelligence analysis. http://www.rand.org/content/dam/
rand/pubs/monographs/2008/RAND_MG668.pdf. Last accessed 18 Oct 2015.
Levitt, S. D., & List, J. A. (2007). What do laboratory experimentes measuring social preferences
reveal about the real world? The Journal of Economic Perspectives, 21(2), 153–174.
Levy, J. S. (1984). The offensive/defensive balance of military technology: A theoretical and
historical analysis. International Studies Quarterly, 28(2), 219–238.
Lewis, D. (2010). Fire strike 7/9. London: Ebury Press.
Long, G. (2014). The edge of glory: The western way of combat and the search for the elusive
decisive battle in an age of terror. Journal of Military Operations, 2(1), 13–16.
Lovelace, Jr. D. C. (2014). Terrorism: Commentary on security documents. Assessing the
reorientation of U.S. National Security strategy toward the Asia-Pacific (Vol. 136). New York:
Oxford University Press.
Luttwak, E. (1987). Strategy: The logic of war and peace. Cambridge: Harvard University Press.
MacGregor, D. (2009). Warrior’s rage: The great tank battle of 73 easting. Annapolis: Naval
Institute Press.
Machiavelli, N. (1984). The prince. Oxford: Oxford University Press. Transl. P. Bondanella.
Machiavelli, N. (1989). Florentine histories. Durham/London: Duke University Press. Transl. A.
Gilbert.
Mack, A. J. R. (1975). Why big nations lose small wars: The politics of asymmetric conflict. World
Politics, 27(2), 175–200.
Markowski, S., & Hall, P. (2011). The reprivatization of war. In D. Braddon, & K. Hartley (Eds.),
Handbook on the economics of conflict (pp. 479–505). Cheltenham: Elgar.
Martínez, L. and J. Schimmelpfennig (2015a). Cómo combatir a los insurgentes. Air and Space
Power Journal 27(2), 32–38.
Martínez, L., & Schimmelpfennig, J. (2015b). The war against the taliban: Tactical operations and
strategic moves. International Journal of Strategic Decision Sciences, 6(3), 18–27.
Maskin, E. S. (2008). Mechanism design: How to implement social goals. The American Economic
Review, 98(3), 567–576.
McCarthy, C. J. (2010). Anti-access/area denial: The evolution of modern warfare. https://www.
usnwc.edu/Lucent/OpenPdf.aspx?id=95. Last accessed 15 July 2015.
McKinstry, L. (2014). Operation sealion: How Britain crushed the german war machine’s dreams
of invasion in 1940. London: John Murray.
Meyerson, R. B. (1991). Game theory: Analysis of conflict. Cambridge: Harvard University Press.
Mitchell, W. (1930). Skways: A book on modern aeronautics. Philadelphia: J.B. Lippincott Co.
Morris, T. S., VanDriel, M., Dries, B., Perdew, J. C., Schulz, R. H., & Jacobsen, K. E. (2015).
Securing operational access: Evolving the air-sea battle concept. http://nationalinterest.org/
feature/securing-operational-access-evolving-the-air-sea-battle-12219. Last accessed 18 Oct
2015.
Mueller, K. P., Castillo, J. J., Morgan, F. E., Pegahi, N., & Rosen, B. (2006). Striking first
preemptive and preventive attack in U.S. national security policy. Santa Monica: RAND
Corporation.
Munkler, H. (2005). The new wars. Malden: Polity Press.
Nagl, J. (2002). Counterinsurgency lessons from Malaya and Vietnam - learning to eat soup with
a knife. Westport: Praeger.
NATO Standardization Agency (2013). Nato glossary of terms and definitions (English and
French), aap-06(2013). http://www.dtic.mil/doctrine/doctrine/other/aap6.pdf. Last accessed 16
Oct 2015.
100 References

Naveh, S. (1997). In pursuit of military excellence: The evolution of operational theory. London:
Frank Cass.
Needham, J., Ping-Yu, H., Gwei-Djen, L., & Ling, W. (1986). Science and civilization in China.
Military Technology; The gunpowder epic. (Vol. 5, Part 7). Cambridge, Cambridge University
Press.
Nussbaum, A. (1943). Just war — a legal concept? MicMichigan Law Re, 42, 453–479.
Nutter, J. (1994). Unpacking threat: A conconcept and formal analysis. In N. A. Graham (Ed.),
Seeking security and development: the impact of military spending and arms transfers (pp. 29–
49). Boulder/London: Lynne Rienner Pub.
Office of Naval History. (1945–1946). United States Naval Administration in World War II: Mine
warfare in the naval establishment. Washington, DC: Office of the Chief of Naval Operations.
Østby, G. (2008). Polarization, horizontal inequalities and violent civil conflict. Journal of Peace
Research, 25(2), 143–162.
Owen, W. F. (2011). Killing your way to control. The British Army Review, (151), 34–37.
Owen, W. F. (2014). The tactics gap: Why we wrestle with the basics. Journal of Military
Operations, 2(1), 17–19.
Paul, T. (1994). Asymmetric conflicts: War initiation by weaker powers. New York: Cambridge
University Press.
Pearce, D. G. (1984). Rationalizable strategic behaviour and the problem of perfection. Economet-
rica, 52(4), 1029–1050.
Powell, R. (2002). Bargaining theory and international conflict. Annual Review of Political
Science, 5, 1–30.
Powell, R. (2004). Bargaining and learning while fighting. American Journal of Political Sci-
ence, 48(2), 344–361.
Quinlivan, J. T. (1995). Force requirements in stability operations. Parameters, XXV(4), 59–69.
Rabin, M. (1993). Incorporating fairness into game theory and economics. The American Economic
Review, 83(5), 1281–1302.
Randolph, S. P. (2015). Strategic decision making and competitive advantage: America vs. north
Vietnam in linebackers I and II. Paper presented at the Violent Skies Symposium, October
15–16, 2015.
Rapoport, A. (1974). Game Theory as a Theory of Conflict Resolution. Dordrecht: D. Reidel
Publishing Company.
Rice, C. (2005). Secretary Condoleezza Rice opening remarks before the senate foreign relations
committee, Washington, D.C. Online on the internet; http://2001-2009.state.gov/secretary/rm/
2005/55303.htm. Last accessed 17 Oct 2015.
Rogerson, W. P. (1984). A note on the incentive for a monopolist to increase fixed costs as a barrier
to entry. The Quarterly Journal of Economics, 99(2), 399–402.
Rosen, S. (1972). War power and the willigness to suffer. In B. M. Russet (Ed.), Peace, war, and
numbers (pp. 167–183). Beverly Hills: Sage Publications.
Rosendorff, B. P., & T. Sandler (2005). Too much of a good thing? The proactive response
dilemma. Journal of Conflict Resolution, 48(5), 657–671.
Salop, S. C., & D. T. Scheffman (1983). Raising rivals’ costs. The American Economic
Review, 73(2), 267–271.
Salop, S. C., & Scheffman, D. T. (1987). Cost-raising strategies. The Journal of Industrial
Economics, 36(1), 19–34.
Schelling, T. C. (1960). The strategy of conflict. Cambridge: Harvard University Press.
Schwartz, N. A. (2011). Airpower in couter insurgency and stability operations. PRISM, 2(Issue
2), 127–134.
Selten, R. (1975). Reexamination of the perfectness concept for equilibrium points in extensive
games. International Journal of Game Theory, 4, 25–55.
Selten, R. (1998a). Aspirational adaption theory. Journal of Mathematical Psychology, 42, 191–
214.
Selten, R. (1998b). Features of experimentaly observed bounded rationality. European Economic
Review, 42, 413–436.
References 101

Shulimson, J., Blasiol, L. A., Smith, C. R., & Dawson, D. A. (1997). U.S. marines in Vietnam: The
defining year, 1968. Washington, D.C.: History and Museums Division, Headquarters, U.S.
Marine Corps.
Simon, H. (1957). Models of man. New York: Wiley & Sons.
Smith, R. (2007). The utility of force: The art of war in the modern age. New York: Alfred Knopf.
Snyder, G. H. (1971). ‘Prisioner’s dilemma’ and ‘chicken’: Models in international politics.
International Studies Quarterly, 15(1), 66–103.
Special Inspector General for Iraq Reconstruction (2010). Quarterly report to the united states
congress. http://www.globalsecurity.org/military/library/report/sigir/sigir-2010-04_report.pdf.
Last accessed 17 Oct 2015.
Steele, D. (2001). Dust, digits and steel: Launching warfare’s future. Army Magazine, 51(6), 23–36.
Stewart, F. (2000). Crisis prevention: Trackling horizontal inequalities. Oxford Development
Studies, 23(3), 245–262.
Strachan, H. (2007). Clausewitz and the dialectics of war. In H. Strachan, & Herbert-Rothe, A.
(Eds.), Clausewitz in the twenty-first century (pp. 14–44). Oxford: Oxford University Press.
Swaine, M. D. (2015a). Beyond American predominance in the wester pacific: The need for a stable
U.S.–China balance of power. http://carnegieendowment.org/2015/04/20/beyond-american-
predominance-in-western-pacific-need-for-stable-u.s.-china-balance-of-power. Last accessed
18 Oct 2015.
Swaine, M. D. (2015b). The real challenge in the pacific. Foreign Affairs, 94, 145.
Tangredi, S. J. (2013). Anti-access warfare: Countering A2/AD strategies. Annapolis: Naval
Institute Press.
Terris, L. G., & Maoz, Z. (2005). Rational mediation: A theory and a test. Journal of Peace
Research, 42(5), 563–583.
Thaler, R. H. (1988). Anomalies: The ultimatum game. The Journal of Economic Perspec-
tives, 2(4), 195–206.
The New York Times (2002). A nation challenged: The economics; attacks may cost U.S.
1.8 million jobs. http://www.nytimes.com/2002/01/13/us/a-nation-challenged-the-economics-
attacks-may-cost-us-1.8-million-jobs.html. Last accessed 18 Oct 2015.
Thiel, J. (2011). The statistical irrelevance of American sigact data: Iraq surge analysis reveals
reality. Small Wars Journal. http://smallwarsjournal.com/jrnl/art/the-statistical-irrelevance-of-
american-sigact-data-iraq-surge-analysis-reveals-reality. Last accessed 18 Oct 2015.
Thomson, S. R. (1966). Defeating communist insurgency (6th ed.). London: Chatto & Windus.
Tootal, S. (2009). Danger close. London: Jhon Murray.
Tse-Tung, M. (1937). On guerrilla warfare. New York: Praeger Publishers Inc. Transl. Samuel B.
Griffith.
Turnbull, S. (2002). Fighting ships of the far east (1). Oxford: Osprey Publishing.
Vegetius, F. R. (1993). Epitome of military science. Liverpool: Liverpool University Press. Transl.
N. P. Milner.
Wagner, D. H., Mylander, W., & Sanders, T. J. (Eds.). (1999). Naval operations analysis.
Annapolis: U.S. Naval Institute Press.
Wagner, R. H. (2000). Bargaining and war. American Journal of Political Science, 44(3), 469–484.
Wikileaks (2010). Afghan war diary. https://www.wikileaks.org/afg/. Last accessed 18 Oct 2015.
Williamson, O. (1968). Wage rates as a barrier to entry: The Pennington case. Quarterly Journal
of Economics, 85, 85–116.
Wittman, D. (1979). How a war ends: A rational model approach. The Journal of Conflict
Resolution, 23(3), 743–763.