Based on your research related to the intelligence cycle, as per above, relate the article by Freeman, to the intelligence cycle and how pre-attack stages can be identified through a robust intelligence cycle.
Patterns of Activity within the Pre-Attack Stages of Terrorist Activity
Michael Freeman, David Tucker, Steffen Merten
Introduction
Do terrorist attacks follow an observable pattern in terms of their pre-attack
planning? If there is a pattern, how reliable is it? Can we use this pattern as an indicator
or warning of imminent terrorist activity? More specifically, can we use these indications
and warnings to predict the timing of a future terrorist attack?
To answer these questions, this study will first divide the terrorists’ pre-attack
activities into nine stages: networking, training, general planning, attack-specific
recruitment, financing, operational planning, weapons procurement, logistical
preparation, and operational preparation. With these stages in mind, we will then
examine a range of terrorism events (twenty-one in all) and identify when each of these
stages occurred. Some of the events that will be studied will include attacks on: the USS
Cole, September 11, the LAX airport, Bali in 2002, Madrid in 2004, the Limburg oil
tanker, the USS Sullivans, the embassies in East Africa in 1998, and the USS Kearsage.
These cases provide variation across multiple dimensions. Some are small, while
others are large. Some are maritime attacks, while others are on land or from the air.
Some occur in the Middle East, while others occur in Asia, Europe, and the U.S. All
these variations in cases will allow us to assess the overall pattern of terrorist planning
activities, but will also allow us to break the cases apart and see if there are differences
according to the type of attack.
Our goal is to determine if there is a general timeframe in which each stage
occurs. For example, if terrorist financing usually occurs three to six months before an
attack, this will provide a valuable warning to intelligence analysts and will have
different ramifications that if it occurred two to three days before an attack. Also, we will
be able to assess the general ordering of these stages. Does each stage usually occur in
some sequential pattern or does their order change for each attack? For example, if
financing activities are observed, should intelligence analysts next look for evidence of
weapons procurement or attack-specific recruitment or something else?
This paper will first describe the operational phases and how they are coded.
Next, we will analyze the data to see if any patterns emerge. Our discussion will focus
on the overall pattern, as well how different subsets of the cases have different or similar
patterns. Finally, this paper will address how this research can be used to develop
indications and warnings of terrorist attacks. To do so, we will show how we can predict
(with some confidence) the timing of unrealized terrorist attacks, specifically the Fort Dix
and JFK airport plots.
Phases of Terrorist Activity
Before a terrorist attack occurs, several other things must occur. The individuals
involved must join the group, get trained, plan the attack, acquire finances, weapons, and
other material, and make final preparations or rehearsals for the final attack. We have
created nine different phases that incorporate these activities.1 Each phase is meant to be
1
Pre-attack activities have been broken up into different numbers and types of phases by different scholars.
See, for example, Terrorism in the Twenty-First Century, appendix A
au.af.mil/au/awc/awcgate/army/guidterr/app_a.pdf, US Army Training and Doctrine Command TRADOC
as distinct as possible, however, we recognize that some may overlap or be hard to
differentiate when coding actual cases. The phases of pre-attack activity are:
(1) Networking and Indoctrination – Introduction of cell members and exposure to
radical doctrine through events such as religious instruction, cohabitation,
meetings, and social activities.
(2) Terrorist Training – Participation of cell members in organized terrorist
training activities (often overlaps with (1)).
(3) General Planning – The decision to conduct a terrorist attack and choice of a
general target area or target set (i.e. ships, bars, Americans, soldiers, etc.). Phase
also includes general shopping for potential targets.
(4) Recruitment – Selection or the activation of cell members for a specific
terrorist operation by more senior terrorist elements.
(5) Financing – The collection and allocation of funds for a specific terrorist
attack.
(6) Operational Planning – Selection of the specific target, detailed reconnaissance
of the target, and specific planning for operation (delivery vector, procurement
methods, explosives construction, etc.).
G2, TRADOC Intelligence Support Activity Threats, Fort Leavenworth, Kansas, 15 August 2007; PreIncident Indicators of Terrorist Incidents: The Identification of Behavioral, Geographic, and Temporal
Patterns of Preparatory Conduct, Brent Smith, Kelly Damphousse, Paxton Roberts; Akiva Lorenz, Al
Qaedas Maritime Threat, www.maritimeterrorism.com/wp-content/uploads/2008/01/al-qaedas-maritimethreat.pdf; Raphael Perl, Combating Terrorism: The Challenge of Measuring Effectiveness, CRS Report,
3/12/2007; Center for Policing Terrorism, Stratfor, Terrorist Attack Cycle www.cptmi.org/terrorist_attack_cycle.html; and Dan Somner, The Wheel of Terror, Buzzle.com,
www.buzzle.com/editorials/6-24-2005-72112.asp. We have created our own breakdown of phases because
some of these others include phases detailing the actual attack and aftermath, while others combine several
distinct activities into a single phase. Also, several of the authors do not offer descriptions about what goes
into each phase, leaving the reader to guess based on simply the title of the phase.
(7) Weapons Procurement – The procurement or receipt of materials for the
construction of explosives or weapons used in the attack itself (fertilizer, rockets,
detonators, accelerant, etc.).
(8) Logistical Preparation – Logistical actions taken in preparation for the terrorist
attack including safe house rental, vehicle procurement, document procurement,
electronics purchase, etc.).
(9) Operational Preparation – Physical Preparations for the imminent terrorist
attack including explosives construction, vehicle alteration, specific explosives
training, multimedia preparation/creation, etc.).
Emergent Patterns
We examined and coded twenty-one cases of terrorist attacks. For each case we
looked for evidence of when the terrorists engaged in activities across the nine phases.
The cases were:
Achille Lauro Hijacking- 1985
World Trade Center Bombing-1993
Oklahoma City Bombing-1995
Khobar Towers Bombing- 1996
Dar al Islam, Nairobi AQ Bombings- 1998
Columbine Shooting- 1999
LAX Millenium Attack- 2000
USS Sullivans-2000
USS Cole Bombing-2000
September 11th Attacks- 2001
Richard Reid attack- 2001
MV Limburg Bombing-2002
Dubrovka Theater Siege- Moscow, 2002
Bali I Bombings- 2002
Madrid Train Bombings- 2004
USS Kearsarge Attack- 2005
Bali II Bombings- 2005
London Subway – 2005
Fort Dix, 2007
JFK Airport, 2007
Virginia Tech Shooting, 2007
These cases were not chosen randomly, and so any conclusions must be read as
only tentative. In fact, the cases were chosen based on two criteria. First, we chose cases
for which we expected to be able to gather data on the different phases. Second, we
intentionally added cases to get as much variation as possible, but this variation may not
occur in the same proportions in the overall universe of cases as it does in our dataset.
For example, we chose big cases like the 9/11 attacks, the Madrid train bombings, and
others that we would expect to require more planning, but also smaller attacks like the
LAX Millenium plot and the Richard Reid failed attack, which presumably would require
less planning. We also chose al-Qaeda attacks as well as attacks by other groups. We
included maritime terrorist attacks (Cole, Sullivans, Limburg), and non-maritime attacks.
We included cases of not just successful attacks, but also ones that failed (Sullivans,
Richard Reid, LAX Millenium), and even ones that were thwarted preventatively (Fort
Dix and JFK airport). We even included non-terrorist attacks with the inclusion of the
Columbine and Virginia Tech school shootings. The purpose of intentionally including
cases of different types was to allow us to be able to break these cases apart and see if
they fit the larger, overall patterns.
For all the cases, there are issues of missing or incomplete data. For example, we
know that the Limburg attackers received a few infusions of cash before the attack.
However, because we dont have a firm date, we couldnt include this information in our
dataset. Consequently, for any single case, we cannot plot the timing of all the activities
by phase that we know must have occurred before the attack.
With these issues in mind, we plotted the data on the graph below (Figure 1).
Figure 1
Figure 2 takes the same data and shows a bracket for each phase. The center of the
bracket is the mean and the edges are two standard deviations from the mean.
Figure 2
Figures 3, 4, 5, and 6 present the data in a similar way, except they only include smaller
subsets of the data (al-Qaeda attacks in Figures 3 and 4, and maritime attacks in Figures 5
and 6)
Figure 3
Figure 4
Figure 5
Figure 6
In terms of patterns that emerge, we must first re-iterate that the limited number
of cases and their non-random selection urge caution in interpreting the results of the
research. As a generalization, we can say that the phases follow in what might be
described as an operationally logical pattern. Recruiting and building general operational
capabilities occurs first, often months in advance of a particular operation. Operational
phases connected to a specific operation take place closer to the event, sometimes only
weeks or days before the attack. Sometimes a triggering event unconnected or at least
not directly connected to the individuals preparing for attacks (an arrest or political event,
for example) leads the terrorists to launch an operation that they have been planning.
Overall, the data shows that many of these attacks were in the planning stages for
quite a long time, often taking years or at least many months. Specifically, operational
phases for nine of the attacks we examined unfolded in 6 to 12 months. Seven attacks
took 12 to 24 months to develop. Five took longer than 24 months. Also, the more
attack-specific phases tended to occur closer to the attack itself, but also in a much
shorter time window (almost always within a month or two of the average).
The pattern for all of the cases also held up for subsets of the data. For instance,
as figures 3-6 show, al-Qaeda attacks and maritime attacks were fairly consistent with the
overall pattern, but exceptions occurred.
Although a general pattern emerged, we could not identify any critical pathways.
The phases did not always occur according to the general pattern. If a group had carried
out a previous attack, then the pattern was foreshortened because the group had already
built its capability and so lead times before an attack could be shorter. Likewise, while
there is a general logic to the ordering of the phases, there is no reason to expect that they
must necessarily be in the same order for all cases.
Indications and Warning
While understanding the general pattern of preparation before a terrorist attack is
useful, it is the most valuable if it can be used as an indicator of an upcoming attack.
Specifically, if we assume that a terrorist plot follows the larger pattern (and this is
admittedly a big and perhaps problematic assumption), can we predict when the attack
itself is likely to occur? In our dataset, we have two cases of pre-empted attacks the
Fort Dix plot and the JFK Airport plot, both in 2007 that are illustrative of this point.
Because the plotters were arrested before an attack could take place, we could not plot
them on the same scale as the other attacks (where the x axis is time before the attack).
Instead, the events are plotted chronologically. Both of these cases fit within the overall
pattern developed from the other attacks. In fact, based on the overall patterns found in
other cases, we could estimate that the attacks would likely have occurred 2-4 months or
so after the plotters were arrested. Figure 7 shows the JFK plot and Figure 8 shows the
Fort Dix plot.
Figure 7
Figure 8
Conclusions
In sum, we have collected data on twenty-one different attacks and coded the
information on their pre-attack activities according to our nine phases. We have found
that the cases follow a general pattern, yet there are outliers to the pattern as well as
nothing necessary about the pattern itself. Nevertheless, the robustness of the pattern
across types and scales of attacks shows that it could be used as an indication of the
timing of possible future attacks.
Recent Comments