MODELING ISIL: UNDERSTANDING THE DYNAMICS OF A MODERN
TERRORIST NETWORK
James R. Enos
Department of Systems Engineering
US. Military Academy, West Point, NY 10996
Abstract
The Islamic State of Iraq and the Levant (ISIL) grew out of the insurgency in Iraq and flourished in the
vacuum left after the civil war in Syria and the withdrawal of U.S. forces from Iraq. Its support among
Sunni Arabs sprang from their distrust of the Shia government in Iraq and the persecution of Sunni Iraqis
after the U.S. withdrawal. Their violent tactics are widely televised as the conduct executions almost
every day that become more and more brutal. Although they display the behavior and structure of a
traditional insurgency, they also show characteristics of an emerging state actor, and have demonstrated
their global reach through acts of violence across the globe. Additionally, their cyber presence on the
web and social media enable them to reach thousands of disenfranchised Muslims across the global and
potentially inspire acts of violence by ISIL followers. This paper reviews the relevant literature and
presents a causal loop diagram to model the dynamic behavior behind ISIL’s rise and continued support
across the globe. It goes on to examine the potential impacts of several policy options to combat ISIL
through the use of these causal loop diagrams.
Key Words
Islamic State, Insurgency Analysis, System Dynamics
Introduction
The Islamic State of Iraq and the Levant (ISIL), sometimes referred to as the Islamic State of Iraq
and Syria (ISIS), is a violent extremist organization that rose to power in the aftermath of the civil war in
Syria and the U.S. withdrawal from Iraq in 2011. The group and its leader Abu Bakr al-Baghdadi preach
an extremist view of Islam and their increasingly violent and brutal tactics have spread across a large
portion of Syria and Northern Iraq (Laub & Masters, 2014). ISIL presents a strategic risk to the entire
Middle East region, threatens to destabilize the region, and has demonstrated the ability to plan and
execute attacks well outside its area of influence.
ISIL traces its roots back to al Qaeda and most of their initial followers supported Abu Musab al-
Zarqawi who led the insurgency in Iraq after the 2003 invasion by the United States (Laub & Masters,
2014). It gained support after the U.S. withdrew its forces from Iraq and the Shia led government under
Maliki marginalized the Sunni minority and consolidated governmental power within a select group of
Shia leaders. Additionally, the terrorist group gained support and fighters from the brutal civil war in
Syria, where a majority of its fighters were trained and gained experience fighting against the pro-regime
forces there (Wood, 2015). ISIL’s leader al-Baghdadi, assumed power in 2010 and in July of 2015 he
delivered a Ramadan sermon from the recently seized city of Mosul, Iraq announcing himself as the first
caliph in generations (Wood, 2015). ISIL sought to increase its religious legitimacy with this act and truly
become an emergent state actor versus a traditional insurgency. A majority of ISIL’s funding comes from
illegal activities, including extortion, crime, and smuggling, and in 2014 some estimates placed their
monthly income at approximately $8 million (Laub & Masters, 2014). However, ISIL also demonstrates
behavior similar to an emergent state actor and has developed schools, a legal system, and implemented
Sharia Law from its headquarters in Raqqa, Syria (al-Tamimi, 2015).
By June of 2014, ISIL had gained territory in Syria and Northern Iraq that surpassed the United
Kingdom in size (Wood, 2015). However, Iraqi security forces, Kurdish militia, and Syrian fighters,
assisted by U.S. Special Forces, have begun to regain gain territory from ISIL. Most recently, fighting in
the Iraqi cities of Ramadi and Mosul resulted in significant losses for ISIL; however, they still remain a
major threat to the region (Al-Marashi, 2016). Figure 1 presents a map of the ISIL held territory as of
March 2017 to include the controlled areas, areas where they recently conducted attacks, and the areas
that generally support their cause (Institute for the Study of War, 2017). As shown, their support comes
from a very large area of Northern Iraq and they control major cities in Eastern Syria, to include their
capital of Raqqa.
BYOF WAR
Change to the Map.
Control
Attack
0000
Support
Iraqi Kurdistan,
Figure 1: ISIL Territory, March 2017 (Institute for the Study of War, 2017)
One of the more barbaric aspects of ISIL are the brutal, graphic executions of prisoners that they
display on the internet and social media to inspire both fear of the group and attract recruits to their cause.
Their executions range from mass shootings to beheadings to drowning and burning prisoners alive in
horrific acts of violence against anyone whom they deem a threat. Their targets have ranged from
American journalists to Iraqi Christians to a Jordanian pilot who was shot down over their territory, each
attack insights global outrage against the group. However, at the same time these acts build the reputation
of ISIL as the true practitioner of their view of Sharia law and following the Islamic religion (Wood,
2015). However, these acts generally receive outrage from Muslim populations around the globe and
most Islamic religious leaders have spoken out against ISIL and believe they are not practicing Islam.
Another interesting aspect of ISIL is the global reach of the terrorist organization and their ability
to sponsor attacks abroad as well as inspire their supporters to conduct attacks on their own. One of the
methods with which they do this are their recorded sermons and speeches on the internet that live well
beyond the life of any individual members of ISIL. One such person, Anwar al-Awlaki, an American
born member of al-Qaeda, continues to inspire members of ISIL and supporters of ISIL to conduct attacks
years after the U.S. killed him in a drone strike (Scott, 2015). This type of internet presence and
recruiting methods has resulted in several attacks across the globe including attacks within the United
States in both California and Florida. Additionally, ISIL has been able to conduct sponsored attacks
through cells in Europe to include the deadly Paris Attacks in November of 2015 and the bombing of the
Brussels airport in 2016. This ability to extend their reach beyond their geographic boarders separates
them from most traditional insurgencies and increases the complex dynamics of the terrorist group.
Literature Review
The literature review section focuses on system dynamics as a method and the literature on
insurgencies and counterinsurgencies. System dynamics provides a means to model an insurgency and
potential policies for countering that insurgency or terrorist group. The review includes a brief
background on system dynamics and focuses on the literature behind causal loop diagrams and stocks and
flows. The insurgency portion of the literature review provides a base of knowledge for understanding
how insurgencies operate and potential counter insurgency measures to combat them. Additionally, it
examines several applications of system dynamics to the modeling and policy analysis of
counterinsurgency operations.
System Dynamics
System Dynamics is a methodology to understand the dynamic behavior of systems by exposing
the underlying systems structure and modeling the behavior over time to adjust individuals’ mental
models of the system. Forrester (1961) described the potential for system dynamics to assist decision
makers understand the implications of their policies and help with level management problems. His
book, World Dynamics, describes his model of the world that described the interrelations between
population, capital investment, geographical space, natural resources, pollution, and food production
(Forrester, 1971). Forrester’s (1968) broad work describes a system as “a grouping of parts that operate
together for a common purpose”. He further classifies two types of systems: open systems, in which
external variables affect the system, or closed systems where all variables are internal to the system
(Forrester, 1961).The behavior of a system over time is the dynamics of a system and because of the
system’s underlying structure they are complex and non-linear in nature (Forrester, 1961). This
lexity stems from feedback within the system, time delays between decisions and effects, and the
learning process of the system (Sterman, 2000).
Applications of System Dynamics have provided insights across several domains; including
corporate policy, infectious disease, commodity markets, and drug addiction, and commodity markets
(Forrester, 1971). Companies and consultants have extensively used System Dynamics for managing
large, complex projects with a great deal of success. One area where businesses utilize System Dynamics
is in the development of corporate strategy and analysis of business decisions after a crisis or complex
problem triggers these shifts in business strategy. System Dynamics assists in determining the root cause
of the crisis and identifying potential consequences of alternative courses of action (Sterman, 2000).
Recently, scholars and analysts have used system dynamics models to gaining an understanding of
insurgencies and counter insurgencies (Anderson, 2009).
Causal Loop Diagrams
Causal loop diagrams, key element of system dynamics, are signed diagrams that represent the
feedback structure within a system. Casual loops differ from discrete, event-oriented perspectives of
individual causes and effects in that any cause is an effect in a closed system (Richardson, 1991). In
System Dynamics, the causal loop diagrams indicate that one variable influences another through either
physical or information flows. These diagrams describe the behavior of the system by talking through the
loop to tell the story of the interactions within the system (Meadows, Randers, & Meadows, 2004).
There are two main types of individual feedback loops: reinforcing or positive loops and balancing
or negative loops. Richardson (1991) describes a reinforcing feedback loop as a chain of cause and effect
relationships that amplify a change in any one of the variables. A change to any variable in the loop in
one direction propagates through the loop to change the original variable in the same direction. The
positive loop designation does not necessarily indicate that the loop will have favorable results, quite
often the opposite occurs. Additional when transferred to a stock and flow model, these loops have the
potential to grow exponentially (Meadows, Randers, & Meadows, 2004).
Balancing feedback loops work to diminish the effect of a change in a system and restore balance
to a system (Richardson, 1991). In System Dynamics, balancing feedback loops bring a system back to a
desired level or a level constrained by the system. Forrester (1968) provides a simple inventory control
system an example of a first-order negative feedback loop; when inventories falls below a desired level,
the order rate increases to increase inventories. In this type of loop, the decrease to a variable propagates
through the loop to increase the original variable.
In natural, human, or engineered systems information or material delays often exist which increase
the dynamic complexity of these systems. In material delays conservation of matter applies, so no units
will be lost during the delay, but the delay will cause the inflow to differ from the outflow, thus resulting
in an accumulation (Forrester, 1961). Whereas in information delays, the information may lose value
over time, so the accumulation of information may diminish over time as people make decisions based on
old information (Forrester, Industrial Dynamics, 1961). These delays can have dramatic effects on a
system’s behavior over time and may cause a system to overshoot is goal when the feedback signal is
delayed. If the system overshoots its natural limit, the delay may result in the eventual collapse of the
system if the delay allows the system to continue to grow to a point where it causes irreversible damage
(Meadows, Randers, & Meadows, 2004).
Stocks and Flows
Stocks and flows are the other major component of a system dynamics models that allow for the
simulation of the system over time. A stock, general depicted as a box, is a measurable accumulation of
material or information within a system. A flow feeds or emanates from a stock and is an instantaneous
rate of change, either material or information, between stocks (Forrester, 1961). Mathematically, a model
determines the value of a stock by taking the integral of the combined inflows and outflows given the
initial conditions of the stock. A simple example of a stock and flow is a bathtub, in which faucet acts as
an inflow and the drain as the outflow with the water being the stock inside the tub. When the inflow
from the faucet exceeds the outflow of the drain the tub fills with water. Conversely, if the drain exceeds
the water from the faucet, the bath tub begins to empty (Sterman, 2000).
Insurgencies
This portion of the literature review focuses on understanding insurgencies, terrorist groups, and
the counter-insurgency operations aimed at containing and defeating these groups. It includes some
background on insurgencies to better understand how these groups operate. Additionally, is provides a
brief summary of the U.S. Army’s recently updated manual on counter-insurgencies. Then, it discusses
recent applications of system dynamics models of insurgencies. These models serve as a basis for
developing specific feedback models of ISIL for this paper.
Background
The Army’s Field Manual 3-24, Counterinsurgency, provides an overview of the structure and
causes of insurgencies and presents counterinsurgency techniques to combat modern insurgencies (U.S.
Army, 2006). This manual assists in understanding the dynamic behavior observed in an insurgency and
the interaction between insurgent and pro-governement forces. Kilcullen (2004) describes an insurgency
as a complex system that needs energy, in the form of acts of violence and grievances against the
government, to sustain itself through several feedback structures. His work focuses on the rise of al
Qaeda and the insurgencies in both Iraq and Afghanistan. FM 3-24 argues that the legitimacy of a nation
is the underlying goal in any counterinsurgency and nation states attempt to gain legitimacy by improving
their economic situation, essential services, and security. Additionally, they attempt to minimize the level
of corruption and violence to a level that is culturally acceptable to the people of the nation (U.S. Army,
2006). This aligns with Kilcullen’s perspective on understanding an insurgency’s effectiveness as the
level of corruption and acts of violence that are requirements for an insurgency’s success (Kilcullen,
4
2004). Over time the legitimacy of both the insurgency and the government can change based on the
population’s perception of the security situation and the essential services delivered by the insurgency or
government (U.S. Army, 2006).
System Dynamics Models of Insurgencies
There have been several efforts of applying system dynamics to the modeling of insurgency and
counter-insurgency. One of the more prolific examples of modeling an insurgency was developed by PA
consulting and published in the NY Times. Although the NY Times article focused on the presentation
and use of PowerPoint, the model provided an excellent example of a causal loop diagram for an
insurgency (Bumiller, 2010). Although the overall diagram was incredibly complex, the detailed
breakdown of the components of the insurgency and security situation in Afghanistan provide excellent
analysis of the system structure. More generalized insurgency and counter-insurgency models also
provide a basis for understanding ISIL and the dynamic behavior of the terrorist group. Anderson (2009)
built a system dynamics model of an insurgency that included both the recruitment aspect of an
insurgency and the suppression aspect of a counter-insurgency.
Choucri, et al (2006) built a system dynamics model of state security that assists in the
understanding of the dynamics behind an insurgency. Their model included structure behind the
movement from the pro-government population to dissidents to insurgents. Additionally they modeled
the potential for the insurgent population to be removed by a security force (Choucri, et al., 2006). Other
work built upon these counter insurgency models to examine the effectiveness of reconstruction projects
within a counter-insurgency. These reconstruction projects focused on building popular support for the
government and decreasing the support for the insurgency (Enos, 2012). Additionally, Enos and Farr
(2013) expanded on these previous works to understand the dynamics behind security and reconstruction
efforts in a counter-insurgency. Their work also examined the potential benefits of investing resources in
building partner capacity before a state fails and falls into an insurgency.
More recent work models the regional rise of ISIL and the dynamics behind the recruitment,
financing, territory, and legitimacy of ISIL. This work differed from traditional insurgency models in that
it focused on ISIL as an emerging state actor that not only works to delegitimize the nation state’s
government, but also become a recognized state as well (Clancy, 2016). Clancy’s model focused on
understanding the legitimacy behind ISIL’s control of territory in Iraq and Syria. It models the amount of
territory controlled by ISIL, size of the fighting force, finances, and the percentage of foreign fighters
recruited by ISIL (Clancy, 2016). All of these factors are important for understanding the basis of ISIL in
the region and this paper expands upon this work to understand the dynamics of ISIL’s impact on a global
basis.
Methodology
This section presents the methodology used to depict the underlying system structure and feedback
mechanisms presence in ISIL and their operations. Causal loop diagrams provide the basis for
understanding the underlying feedback structure of ISIL and help to explain the dynamic behavior of the
group. This work does not go as far as to build a functioning model of ISIL, but it does provide some
basic stock and flow structures for ISIL that provide a basis for future modeling efforts. The causal loop
diagrams and stocks and flows serve as structure for the following section’s analysis of potential policy
options to combat ISIL.
System Structure
The overall system structure consists of at least five reinforcing and five balancing loops that create
the dynamic behavior observed of ISIL and its followers. First, a basic reinforcing loop that models the
terrorist groups acts of violence, recruits, and population depicts how the group builds its membership.
This loop is balanced by a loop that depicts the fear and outrage against ISIL which decreases the support
for the group. The acts of violence also call for a traditional counter-insurgency against ISIL when their
actions reach a tipping point with the global community. There are two other reinforcing loops that
relatively new to traditional counter-insurgency models and those depict the social media and web
presence of ISIL. These loops also encourage the global reach of ISIL and ISIL inspired groups to
conduct attacks such as those seen in Paris, Florida, and California. The remainder of this section
discusses each of these loops in greater detail.
Figure 2 depicts the basic insurgency recruitment and acts of
violence feedback mechanism that is present in most insurgencies or
terrorist groups (Anderson, 2009). Designated as R1, for the first
reinforcing feedback loop, this loop begins with an /S/L Population,
which conducts operations that result in Acts of Violence. These acts of
violence add to the /S/L Recruiting Message which targets dissident
populations within the geographical are to bring them into the group.
The intent of these recruiting messages is to increase the number of /S/L
Recruits, which after a training period become members of the JS/L
Population and add to this stock of terrorists.
As evident in the Sunni-Awakening in Iraq, 2006-2008,
insurgencies or terrorist groups run the risk of alienating their population when their acts of violence
increase to a point where the local population will not
+
Acts of
Violence ISIL Recruiting
ISIL +
ISIL Recruits
Population
ys
Figure 2: R1 - ISIL Recruitment
+ — * tolerate their actions. In Iraq, al-Qaeda of Iraq (AQI)
Fear of ISIL Vislonee ISIL Recruiting experienced this in the Anbar province where their acts of
Massage terror in the region turned the local sheiks and tribes against
js i) them, leading to the success of the US led
\ counterinsurgency and removal of AQI from Al Anbar
Support for IsIL + (Doyle, 2011). Figure 3 presents the causal loop for this
IsiL ISIL Recruits
Population
~—_
+ +
Figure 3: B1 - Violence Blowback
portion of the system that shows how terrorist acts of
violence can blow-back against the insurgency in the long
run. As the group increases their Acts of Violence, it also
increases the Fear of ISIL in the local region. As
insurgencies and terrorist groups rely on the local population to house, fund, and support them, this
becomes a problem for most insurgencies (U.S. Army, 2006). If the local population fears ISIL to the
point that they no longer align with the group, the Support for ISIL will decrease and without this support
the group cannot support the /S/L Population stock.
The next diagram describes the traditional counter insurgency feedback that is present in the
majority of counter-insurgency operations. The Acts of Violence trigger a Desire to Defeat ISIL through
the use of counter-insurgency
tactics (Anderson, 2009). As
insurgencies are non-state actors,
+
sit io
Defeat ISIL
+
i B2
some of the tradition means Counter-ist @s
outside of force, like embargos or Operations Ge Se od Yq
sanctions, are not as effective at + Fear of ISIL Violence ISIL Recruiting
. . Message
5 +
countering an insurgency (Enos & Nation Building
Farr, State Security Dynamics and
the Impact of Intervention to
Build Country Capacity, 2013).
As depicted in Figure 4, this loop
begins as the JSIL Population
conducts Acts of Violence within
the region. At some point, this
will trigger a Desire to Defeat
ISIL within local, regional, and
te)
Operations
Support for +
ISIL Recruits
@
IsIL
_
+ tA +
Is
Population
—
Direct Action
Operations
Figure 4: B2 & B3 - Counterinsurgency
eventually global governments that will eventually increase the number of Counter-ISIL Operations.
These operations generally come in two forms, direct military action against the insurgency or nation
building to decrease support for the group. Direct Action Operations target individuals or parts of the
terrorist organization through the use of military force and intend to directly decrease the /S/L Population
(Anderson, 2009). Whereas, Nation Building Operations indirectly decrease the [SIL Population by first
decreasing the Support for ISIL, which in turn decreases the SIL Population variable (Enos, 2012).
An interesting aspect of ISIL is their use of
social media to propagate their message and attract Acts of
new recruits from across the globe to join their cause. Violence ‘SIL anne,
+
They have become very adept at using social media f Moses +
sites to target disenfranchised individuals and recruit Ant) ISIE Social Media
them to their cause in Iraq and Syria (Farwell, 2014). J Ana) a
For this reinforcing feedback loop, as the /S/L isi + ae
Recruiting Message increase, it in turn increases the PePulation _ISIL Recruits
ISIL Social Media Presence. With an increased *
presence on social media, the group is able to attract Figure 5: R2 - ISIL Social Media
new members globally and increase the /S/L Recruits
within Iraq and Syria for training. As Figure 5 depicts, this eventually leads to an increase in the /S/L
Population, more Acts of Violence, and eventually more material for the [SIL Recruiting Message which
helps grow the terrorist group. It is estimated that tens of thousands of recruits from France, the United
Kingdom, Belgium, German, and the United States have entered the Islamic State to become part of the
terrorist organization (Wood, 2015).
One aspect of the fight against ISIL that sets it apart from tradition insurgencies is the movement of
ISIL into the cyberspace domain through the use of social media and the internet. Figure 6 presents
several of the reinforcing and balancing loops that help to explain the dynamics of operating against an
insurgency in the cyber domain. First, the loop designated as R3 depicts how the /S/L Videos increase the
ISIL Social Media Presence, thus resulting in more /S/L Recruits. These recruits then join the /S/L
Population and conduct additional Acts of Violence, resulting in more video content for social media.
These videos also increase the [SIL Web Presence which in turn increases the JS/L Recruits in another
reinforcing feedback loop to increase the JS/L Population. However, these Acts of Violence, especially
the barbaric nature of the videotaped executions, increase the Desire to Defeat ISIL which results in
additional Counter-ISIL Operations. In this case, a new aspect of a counter-insurgency is added to the
model through Cyber Operations which are aimed at decreasing the [SIL Web Presence. In this balancing
feedback mechanism, there is a significant delay from the time content is posted until it can be countered.
+
Desire to + +
Defeat ISIL ISIL Videos
ye
Counter.1SIL
erations + +
°r Acts of
/ * +
| idee A
| ) ISIL Social Media
resence
) :
Is ISIL Web
i Population ISIL Recruits Ana) Presence
. oa ee ne oY
ame
Operations
—
+
Figure 6: R3, R4, and B4, ISIL Cyberspace Domain
ISIL ability to reach well beyond its geographic boundaries is another aspect of the terror group
that makes it unique from some other historical insurgencies. Similar to al-Qaeda, ISIL sponsors terrorist
cells across the globe capable of operating independently to carry out attacks. ISIL conducted a major
attack in Paris in November of 2015 in a coordinated attack across the city in which 130 people were
killed. Additionally, in March of 2016, ISIL cells conducted attacks across Brussels which included
suicide bombers at the airport and a train station (Callimachi, 2016). These acts are examples of ISIL’s
ability to strike terror across the globe. In this loop, the /S/L Population increases the [SIL Sponsored
Attacks Abroad, which in turn create more Acts of Violence credited to the group. This increases the
recruiting and brings additional people into the JS/L Population in a reinforcing feedback loop. However,
by extending their reach beyond their geographic boundaries, they also trigger a balancing feedback loop
as these attacks increase the Global Outrage Against ISIL. Figure 7 presents this feedback loop and
shows how the Global Outrage Against ISIL increases the Desire to Defeat ISIL and calls for additional
Counter-ISIL Operations to include direct action, cyber operations, and nation building.
__ Global Outrage
———*ISIL Videos —
8 \\ \
exact
| Foar of fsit Veet sit Recruiting \ a
. + Me: \
_—- a ee
| Presence . (asp /
ISIL Web
| rations:
Direct Action
ai ~
cyber
Operations
Figure 7: R5 and B5 ISIL Sponsored Attacks Abroad
" Defeat ISIL aa Sis ISIL Insipred
CounterSiL @ i 3 ra
unter a + "i
Orerations —- ~ ro
| ‘ ral "fat eensied “1
Fintan in “f ae fs
| Pr) | ISIL Social Media }
| \ Operations Presence
| i)
i Gee oS wh S
Rime
Operations Gs
Figure 8: B6 ISIL Inspired Attacks
In addition to the sponsored attacks, ISIL has also inspired attacks across the globe by
disenfranchised Muslims who pledge allegiance to the group, but are not actually part of the terror group.
Figure 8 depicts how the /SIL Web Presence increases the number of JS/L Inspired Attacks, but this only
adds to the Global Outrage Against ISIL. Although these inspired attacks may increase recruitment for
ISIL, it is likely that they will create balancing feedback as the Desire to Defeat ISIL is increased with
these attacks, resulting in additional Counter-JSIL Operations. In the San Bernardino attack a husband
and wife who had pledged allegiance to ISIL killed 14 people at a business gathering prior to being killed
by police in gun battle in the streets (Almasy, Lah, & Moya, 2015). In Orlando a shooter who pledged
allegiance to ISIL gunned down 49 people in a nightclub and wounded another 53 before being killed by
Police as they raided the nightclub (Ellis, Fantz, Karimi, & McLaughlin, 2016). These individual
incidences demonstrate another danger aspect of ISIL and their ability to not only plan attacks across the
globe, but also inspire others to conduct similar acts of violence.
Stocks and Flows
This portion of the methodology begins to develop the stock and flow structure as building blocks
for a functioning model of the ISIL terrorist group and their operations. It focuses on three aspects of the
insurgency group and builds on previous work to create three stock and flow models. The first model is
the regional insurgency recruitment model, which models the /S/L Population as described in reinforcing
loop one. The second stock and flow model depicts the /S/L Web Presence and some of the difficulties in
combating this part of the terrorist group.
In the first stock and flow model, the basic insurgency model includes the Pro-Government
Population which has the potential to become part of the Dissident Population if the level of security and
support from the government decreases. Figure 9 presents this stock and flow model which would be
used as a building block for a more detailed model of ISIL and their activities. Additionally, the
Dissident Population could return as the government increases its legitimacy by providing basic needs for
the population. However, through the /S/L Recruiting Message, which targets the Dissident Population,
ISIL is able to gain recruits into their organization. After a training period, they become part of the /S/L
Population, capable of carrying out Acts of Violence against the government and pro-government forces.
However, this violence increases the Desire to Defeat ISIL and thus increases the Direct Action
Operations with remove terrorists by killing or capturing members of ISIL.
ISI. Recrating ets of —* prac Action
“Message - Operations
i }
= [ }
ne {
Prosovernment | | | i ] —_—
Population ISI. Rooraite $——r si Popuition meena ae
| Rate of | | Rateor | [Fate ortorrnat | Terrorists
Recruiting ‘Training Removal
___Nation Building
~ Operations
Figure 9: Basic ISIL Recruitment Model
One of the unique aspects of ISIL is its web presence and the use of both the internet and social
media to attract recruits from across the globe. Figure 10 presents the overall structure of the cyber
domain for ISIL. First, the Acts of Violence often result in ISIL Videos, which include both offensive
operations on the ground and the execution of prisoners by ISIL operatives. These videos then become
part of the active ISIL Web Presence that is used to recruit additional members to the organization. After
a period of time, as with any web content, some of the presence becomes inactive; however, the
possibility to access this content remains. Only through the use of aggressive Cyber Operations can the
pro-government forces hope to remove the ISIL Web Presence. Similar to the removal of the terrorists
9
themselves, the web content could also be removed. However, there are several challenges associated
with his aspect of a Counter-ISIL Operations. First there is a large time delay between the time a video is
posted and when it can be targeted. Also, it may be difficult to fully remove the presence form the
internet as the content may exist on several different servers and even individual nodes of the internet.
<Acts of Violence>
ISIL Videos
Rate of Web
Content ity
10: ISIL Cyberdomain Presence
Analysis
This section uses the causal loops developed to determine the potential impact of several different
policy options to combat ISIL on a global front. The first policy is to treat ISIL like a traditional
insurgency and focus efforts on defeating ISIL on the ground in Iraq and Syria. The second policy looks
at the possibility of pursuing an approached based on capacity and nation building within Iraq and Syria
to gain support for the governments of those nations and pull support away from ISIL. The third
approach focuses on attacking ISIL in the cyber domain to degrade their message on the internet and
remove their presence from social media. Finally, an approach that combines a heavy direct action
counter insurgency and cyber operations is examined.
The first approach follows traditional counter-insurgency tactics and attempts to decrease the /S/L
Population through Direct Action Operations. These would be targeted operations against individual
leadership of ISIL or attacking strongholds of ISIL members across Iraq and Syria. By directly attacking
the /SIL Population, the pro-government forces hope to stop the Acts of Violence by taking advantage of
the balancing aspects of the counter-insurgency feedback loop, designated as B3 in the previous figures.
However, they would have to combat the other reinforcing feedback loops present that support the growth
of ISIL in order to reach a tipping point where this approach would result in the collapse of ISIL.
The second approach indirectly targets the /S/L Population through the use of Nation Building
Operations depicted in loop B2. In this approach the Counter-JSIL Operations would focus on the
humanitarian side of counter-insurgency in order to decrease the Support for ISIL. As Killcullen (2004)
points out, the lifeblood of an insurgency is the support that it receives from the local population. With a
decrease in Support for ISIL, the ISIL Population would not be able to sustain itself and would eventually
die out. However, this assumes that ISIL would not look elsewhere for support or potentially turn their
brutal tactics against the population to a point where they fear and support ISIL more than the pro-
government forces.
However, both of these approaches focus on the regional threat posed by ISIL in Iraq and Syria and
do not address the global reach of the organization. In the third policy takes a cyber approach to
combating ISIL in the cyber domain to directly target the /S/L Web Presence and ISIL Social Media
Presence. Through an increase in Cyber Operations, pro-government forces would decrease the [SIL
Web Presence as feedback loop B4 indicates. However, there are several problems with this type of
approach. First, the delay between ISIL developing web content and the counter-ISIL forces being able to
find, target, and remove this content. Additionally, it is incredibly difficult to truly remove web content
10
as it might exist in multiple places. Take for example the sermons of Anwar al-Awlaki, which exist ten
years after his death and continue to inspire attacks (Scott, 2015).
The final approach looks to use the entire spectrum of Counter-ISIL Operations in order to take
d ge of all of the balancing feedback loops in the model. Pro-government forces must participate in
Direct Action Operations to remove key leadership of ISIL and retake the territory seized by ISIL in Iraq
and Syria and return it to the rightful control of these states. At the same time, forces must work on
Nation Building Operations, as the security situation dictates, in order to decrease the Support of ISIL in
the local geographical regions. Finally, by incorporating a robust, offensive cyber war against ISIL, the
Cyber Operations would be able to decrease the /S/L Social Media Presence and ISIL Web Presence in
order to decrease the global reach of the organization. This is the only method to deal with both the
regional aspects of ISIL and the /S/L Inspired Attacks across the globe. The other aspect of this policy is
that it cannot end with the defeat of ISIL, as another group may rise in its place. The Nation Building
Operations must continue until both Syria and Iraq are fully capable of stability in the region and
preventing the next al Qaeda or ISIL from emerging.
Conclusion and Future Work
The purpose of this paper is to present a basic understanding of the system structure behind the
terrorist group ISIL. It focuses on understanding the feedback mechanisms that exist in ISIL that result in
the dynamic behavior observed and presents some of the differences between ISIL and a traditional
insurgency. It goes on to create a few stock and flow diagrams that will be the building blocks for future
work to develop a functioning system dynamics model of the terrorist group. As described, it examines
several potential policy options at a high level using the causal loops developed as part of the analysis of
ISIL’s structure. Based on the analysis of these loops, it is likely that only a combined cyber and direct
action approach will truly remove ISIL as a global threat.
There is still much future work to be done in order to make this a truly useful model of ISIL and
serve as a base model for modern insurgencies. Although it moves beyond tradition insurgency models
by incorporating the social media and internet aspects of a modern insurgency it does not have a function
model of the terrorist group and its actions. Future work should include building a system dynamics
model of ISIL, calibrating this model to know data about the size of the group, territory it claims, and acts
of violence. The model could then be used to better understand how the group has been able to attract so
many foreign fighters to its ranks and inspire acts of terror across the globe. With a calibrated model,
more detailed policy analysis would be possible and a more solid recommendation could be made for how
to combat ISIL on all fronts.
It is widely accepted that ISIL must be removed from the globe as it threatens the security of world;
however, there are several different approaches that may defeat ISIL. It is important to understand the
potential unintended consequences from this action. Is it possible to defeat ISIL and remove their
message from the globe? Or will the actions of a few nations in an effort to defeat ISIL just create an
even worse enemy of the world? Hopefully, through the use of system dynamics modeling, these
questions can be answered and more information will assist decision makers understand the consequences
of their actions or inactions.
References
Al-Marashi, I. (2016, January 3). How Iraq recaptured Ramadi and why it matters. Al Jazeera.
Almasy, S., Lah, K., & Moya, A. (2015, December 3). At least 14 people killed in shooting in San
Bernardino; suspect identified. CNN.
al-Tamimi, A. J. (2015). The Dawnof the Islamic State of Iraq and ash-Sham. Current Trends in Islamic
Ideology, 6.
Anderson, E. G. (2009). Modeling Insurgencies and Counterinsurgencies. Austin, TX.
Bumiller, E. (2010, April 26). We Have Met the Enemy and He is PowerPoint. New York Times.
Callimachi, R. (2016, November 22). ISIS Operative Suspected in Paris and Brussels Attacks Is
Identified. NY Times.
Choucri, N., Electris, C., Goldsmith, D., Mistree, D., Madnick, S., Morrison, J. B., . . . Sweitzer-
Hamilton, M. (2006). Understanding & Modeling State Stability: Exploiting System Dynamics.
IEEE Aerospace Conference. IEEE.
Clancy, T. (2016). The Dynamics of ISIS — An Emerging State Actor. 34th International Conference of
the System Dynamics Society. Delft, Netherlands.
Doyle, W. (2011). A Soldier's Dream: Captain Travis Patriquin and the Awakening of Iraq. New York,
NY: NAL Caliber.
Ellis, R., Fantz, A., Karimi, F., & McLaughlin, E. C. (2016, June 13). Orlando shooting: 49 killed,
shooter pledged ISIS allegiance. CNN.
Enos, J. R. (2012). Dynamics of Reconstruction Projects. 30th International Conference of the System
Dynamics Society. St. Gallen, Switzerland.
Enos, J. R., & Farr, J. V. (2013). State Security Dynamics and the Impact of Intervention to Build
Country Capacity. 3/st International Conference of the System Dynamics Society. Cambridge,
MA.
Farwell, J. P. (2014). The Media Strategy of ISIS. Survival: Global Politics and Strategy, 56(6), 49-55.
Forrester, J. W. (1961). Industrial Dynamics. Waltham, MA: Pegasus Communications Inc.
Forrester, J. W. (1968). Principles of Systems. Cambridge, MA: Wright-Allen Press, Inc.
Forrester, J. W. (1971). World Dynamics. Cambridge, MA: Wright-Allen Press, Inc.
Institute for the Study of War. (2017, March 9). SIS Sanctuary Map: March 9, 2017. Retrieved from
Institute for the Study of War: http://www.understandingwar.org/backgrounder/isis-sanctuary-
map-march-9-2017
Kilcullen, D. (2004). Countering Global Insurgency.
Laub, Z., & Masters, J. (2014). The Islamic State in Iraq and Syria. Assyrian International News Agency.
Meadows, D., Randers, J., & Meadows, D. (2004). Limits to Growth: The 30-year Update. White River
Junction, VT: Chelsea Green Publishing Company.
Richardson, G. P. (1991). Feedback Thought in Social Sciences and Systems Theory. Philadelphia, PA:
University of Pennsylvania Press.
Scott, S. (2015, August 27). The Lessons of Anwar al-Awlaki. NY Times.
Sterman, J. (2000). Business Dynamics: Systems Thinking and Modeling for a Complex World. Boston,
MA: Irwin McGraw-Hill.
U.S. Army. (2006). FM 3-24 Counterinsurgency. Washington, D.C.: Department of the Army.
Wood, G. (2015, March). What ISIS Really Wants. The Atlantic.
About the Author
Lieutenant Colonel James Enos is currently a PhD Student at Stevens Institute of Technology in the
Systems Engineering program. During his 17-year Army career, he served as a Systems Engineer on the
Joint Staff and an Assistant Professor in the Department of Systems Engineering at the United States
Military Academy (USMA), West Point, NY. Additionally, he has held numerous leadership positions as
an infantry officer, including Rifle Company Commander, Ranger Instructor, and Platoon Leader. He
graduated from the USMA in 2000 with a Bachelor of Science degree in Engineering Management. He
earned his Master's of Science in Engineering and Management in 2009 from the Systems Design and
Management program at MIT. Upon completion of his PhD, planned for 2018, he will return to the
Department of Systems Engineering at the USMA as an assistant professor.
