cothren indicators twg102006

No description available.

Pre-Incident Indicators of Terrorism Events: Pre-Incident Indicators of Terrorism Events Portions of this research were funded by the National Institute of Justice (Grant Number 2003-DT-CX-0003) and the National Memorial Institute for the Prevention of Terrorism in Oklahoma City (Grant Number MIPT 106-113-2000-064) through the Department of Justice and the Department of Homeland Security. The opinions presented here do not represent the official position of the DOJ, the DHS, the MIPT, or the NIJ. Brent L. Smith University of Arkansas Kelly R. Damphousse University of Oklahoma Paxton Roberts University of Arkansas Terrorism Research Center in Fulbright College, University of Arkansas Jackson Cothren University of Arkansas

Differences in Traditional Crime and Terrorism: Differences in Traditional Crime and Terrorism Primary/Immediate Victims Pecuniary/Heat of Passion Motives Spontaneity No Preparatory Crimes Secondary/Instrumental Targets Political Motives Considerable Planning Series of Preparatory Crimes Terrorism Traditional Criminality

Major Findings of the American Terrorism Study: Major Findings of the American Terrorism Study Terrorist group members engage in significant criminal conduct en route to committing a terrorist incident. They are learning from us, just as we learn about them. They modify their behavior to avoid FBI terrorism investigations.

Terrorist Tactics That Emerged in the Early 1990s: Terrorist Tactics That Emerged in the Early 1990s Leaderless resistance Use of the internet Website “hit lists” of potential targets “Fatwahs” All of which gave rise to “lone wolves” and “elves”

Despite These Changes: Despite These Changes A number of preparatory crimes have to occur to commit most terrorist incidents. They may be committed by fewer group members and fewer meetings may occur: But that may allow local law enforcement to be more easily alerted to the activities of a single individual as opposed to a group which may disperse the preparatory duties. If these behaviors are routinized, patterns of conduct may be identified that might allow early intervention.

Terrorism in Time and Space: Terrorism in Time and Space These behaviors probably occur in routinized temporal and spatial patterns. What are these patterns? Even the most rudimentary questions about terrorist group planning activities have not been addressed.

Background: Background The American Terrorism Study began in 1988 in collaboration with the FBI’s Terrorist Research and Analytical Center. The FBI compiled the names and court case numbers of persons indicted since 1980 to begin the project.

Authorization: Authorization Sponsorship: Since the mid-1990s, the project has been “sponsored” by the House of Representatives Judiciary Subcommittee on Crime and the Senate Judiciary Committee. Legal Review: Release of the data has been approved by the FBI Office of General Counsel and Congressional review.

Methodology: Methodology Utilizes “official” FBI measures of terrorism. Thereby, accepting: The FBI “academic” definition of terrorism The AG regulatory definition when applied

FBI Definition of Terrorism:                     FBI Definition of Terrorism The unlawful use of force or violence against persons or property to intimidate or coerce a government, the civilian population, or any segment thereof, in furtherance of political or social objectives.

FBI Regulatory Definition of Terrorism: FBI Regulatory Definition of Terrorism Dictated and put into practice by: For Domestic Terrorism: The Attorney General Guidelines on General Crimes, Racketeering Enterprises, and Domestic Security/Terrorism Investigations and subsequent editions (1983, 1989, 2002). For International Terrorism: The Attorney General Guidelines for FBI Foreign Collection and Foreign Counterintelligence Investigations.

Characteristics of the ATS Databases: Characteristics of the ATS Databases Currently includes information on: 9,071 federal criminal counts 641 “indictees” 575 persons 71 terrorist groups Representing approximately 90 percent of the population of indicted terrorists during for the period 1980-August 15, 2002 and approximately 50 percent of indicted terrorists for the period August 16, 2002 – August 15, 2004.

Number of Persons and Cases: Number of Persons and Cases

Data Collection: Data Collection Information on 75 variables: demographic and group characteristics, target information, count and case outcomes, conviction rates, sentencing data. Court case documents: dockets, indictments, motions, judgment orders. Temporal and spatial data: 285 variables relating to addresses of terrorists and targets, locations of preparatory crimes, and sequencing of events. Summary of each case

Oracle Database Schema: Oracle Database Schema D _ PERSON PK ID TYPE _ PERSON LAST FIRST MIDDLE SUFFIX BORN DEATH YEARBORN SEX RACE ETHNICITY EDUCATION OCCUPATIO INCOME MARITAL CITIZEN BORNCTRY BORN _ ST BORN _ CTY BORNCITY MIL _ TIME BRANCH EXPERTISE INC _ BEGIN INC _ END ROLE CELL _ ST CELL _ END CELL _ NAM GROUPNAM AG _ MEM AG _ FINAN RECRUITED MILITARY CRIM _ REC ARRESTED INCARCER COMMENT 1 COMMENT 2 COURTCASE CELL _ ST _ M CELL _ ST _ D CELL _ ST _ Y CELL _ END _ M CELL _ END _ D CELL _ END _ Y E _ PERSON PERSON _ ID EVENT _ ID E _ CASE _ STUDY D _ ACCESS RELIABLE VALID SUBJECT COURT PLAIN DEFEN CASE _ NUM DOC _ TYPE COMMENTS BODY _ CITATION SYN SME URL

Proposed Structure: Proposed Structure Pre-incident Data Spatial and Temporal Court Case Data Demographic and Group Post-Indictment Data Prosecutorial & Defense strategies Case and Count Outcomes Statistical database Spatial Information (point records) Court documents Related articles & references Case narratives Web-based Oracle Spatial Database

Slide17: The American Terrorism Study & The Terrorism Knowledge Base (www.tkb.org)

Slide19: Searching by names

Slide24: Searching by keyword

Pre-Incident Indicators Project Objectives: Pre-Incident Indicators Project Objectives Can we identify spatial patterns of antecedent criminal conduct that will increase the probability of successful early intervention before actual terrorist incidents occur? What are the temporal dimensions of terrorist group planning? How many activities are typically involved? And do these activities vary by group type?

Major Temporal and Spatial Dimensions of Terrorism: Major Temporal and Spatial Dimensions of Terrorism Planning and Preliminary Organization   Identification and clarification of roles in group   Exposure to terrorism training manuals   Discussion of potential targets   Drawing schematics   Assignment of tasks  Preparatory Activities Theft of explosives, detonation cord, or weapons   Counterfeiting/ laundering money to support group activities   Procurement of false IDs   Modifications to semi-automatic weapons to fire automatic   Buying of ANFO or other bomb materials such as timers and electrical components   Establishment of bomb making labs Terrorist Incident   Bombings   Assassinations   “Ecotage”   Hostage taking   Hoaxes   Threats   Hijackings   Recruitment   Membership in extremist group   Attendance at rallies or demonstrations   Exposure to Internet sites   Personal recruitment   Accessing extremist literature T1 Time T2 T3 T4 Time T5 T6 T7 D1 Distance/Location D3 D4, D5 D2 Distance/Location D4, D5 D6   Establishment of Residences    Residential Locations May be correlated with other geo-political or census data        

Project Methods for NIJ Project: Pre-Incident Indicators of Terrorism Activities : Project Methods for NIJ Project: Pre-Incident Indicators of Terrorism Activities

Project Methods for Current NIJ Project: Project Methods for Current NIJ Project Select case studies for collection of data. For each case study, identify one or more terrorist incidents for collection of spatial and temporal data (not necessarily an exhaustive or even unbiased sample). Collect data using open source documents. Input information into (first Access, then) Oracle 10g database. Geo-code incidents to create spatial data using ESRI’s ArcGIS suite. Conduct spatial and temporal analysis by leveraging the power of the geodatabase using Intergraph’s GeoMedia Professional, ESRI’s ArcGIS, and Matlab. Provide trainers with a tool for educating law enforcement personnel on patterns of terrorist activities and provide researchers with the database for further analysis.

Case Study Selection: Collected 63 case studies for the period 1980 to 2002 from the four main terrorism group types using open source data Sources for selecting case studies: 1. American Terrorism Database contained information from 1980- mid 1998 at that time 2. List from FBI sent to ATS added period: mid 1998- August 15, 2002 3. Suggestions by members of research team and consultants Case Study Selection

Slide38: Oklahoma Constitutional Militia Order- 2 Phineas Priests Ruby Ridge Seace Conspiracy Third Continental Congress Up the IRS- 3 Washington State Militia White Patriot Party Woodring Homrich El Rukns EPB-Macheteros FALN- 4 May 19 Communist Order- 3 New African Freedom Fighters Nyack Armored Car Robbery United Freedom Front Yahweh Left Wing Right Wing Aryan Nations Aryan Peoples Republic Bixby SC Covenant Sword & Arm of the Lord Felton Chase Jewish Defense League KKK- 3 Krar IDC Minnesota Patriots Council Ohio Unorganized Militia Oklahoma City Bombing Single Issue International Coronado MSU Arson Clayton Lee Wagner Dr. Robert Goldstein EcoRaider ELF Long Island Arsons Ellerman Utah Bombing Fairfield Snow Bowl Free Critter Eugene Arson Griffin Florida Assassination Hill Florida Assassinations Kopp Amherst Assassination Nebraska Golf Vandals Santa Cruz 2 Sherman Oregon Firebombing- 2 Tucson Vandals Unabomber Vance Assassination Wisconsin mink release Abu Nidal Organization Hezbollah Japanese Red Army Lackawanna Six Millennium Conspiracy New York City Conspiracy 1994 New York Subway Bombing Omega 7 Provisional Irish Republican Army- 2 World Trade Center Bombing 1993

Variables: Variables Demographic- age, sex, education, occupation, income, etc. Spatial- address, city, state, zip Temporal- hour, day, week, month, year Miscellaneous- target type, incident damages & deaths, aliases

Slide41: CIRAT Database Development

Slide42: CIRAT Database Development

Slide43: CIRAT Database Development

Data Collection Results: Data Collection Results Records are geocoded using street address and sometimes zip codes. ESRI StreetMap data used for geocoding (this round). Accuracy is currently be assessed independently from a sufficiently large sample.

Temporal Patterns: Temporal Patterns How long is it from the time members are recruited into a terrorist cell before they actually begin conceiving and planning for an incident? How long do terrorist groups plan and prepare before committing a terrorist incident? How many, and what types of, crimes do they commit while preparing for a terrorist incident?

Temporal Patterns: When Recruited Cell Origination Unknown due to small sample size Unknown due to small sample size Terrorist Incident avg. 99 / *54 days n = 191 Planning and Preparatory Behaviors Temporal Patterns n=34 /*n=30 excludes outliers 1st Known Activity avg. 1205 days (3.3 years) avg. 58 / *12 days avg. 41 / *25 days

Temporal Analysis: Earth Liberation Front: Temporal Analysis: Earth Liberation Front

Slide48: Temporal Analysis: Phineas Priests March 30, 1996 through October 8, 1996

Temporal Analysis: NYC Conspiracy: Temporal Analysis: NYC Conspiracy January 3, 1990 through June 24, 1993

Slide50: Breakdown of Incidents by Day of Week n = 236

Slide51: Breakdown of Incidents by Time of Day n = 57

General Spatial Patterns: General Spatial Patterns Conventional criminals typically commit their crimes within five miles of their place of residence. Sources: Thomas Reppetto, Residential Crime, 1974; Richard Wright and Scott Decker, Burglars on the Job, 1996; Richard Wright and Scott Decker, Armed Robbers in Action, 1997. Do terrorists reflect similar patterns of behavior?

General Spatial Patterns: General Spatial Patterns Major terrorist incidents in the U.S. typically involved perpetrators from places other than the site of the target. Timothy McVeigh Oklahoma City, OK Eric Rudolph Birmingham, AL September 11 Attackers New York City, NY

Slide54: Issues of the variation in scale of data of the locations for incidents, preparatory acts, residences etc. can make it difficult for analyzing linear distance between locations. When locations could only be identified down to a zip code level, analysis was conducted only if the margin of error for location within the area was less than 10% of the total distance between the two points being measured. city zip address Locations have been geocoded down to at least the Zip Code, Zip+4, or address level. Geocoding and Distance Analysis

Distances from Terrorist Residences to Incident Locations: Distances from Terrorist Residences to Incident Locations 45% are 0-30 miles 7% are 31-90 miles 8% are 91-270 miles 12% are 271-810 miles 28% are 811-2570 miles Graphed Ranges Median: 394 Median: 79 Mode: 0 Std.dev: 497 n = 433

Distances from Terrorist Residences to Incident Locations: Mean: 283 miles Mode: 0 miles Median: 102 miles Right Wing Mean: 458 miles Mode: 0 miles Median: 48 miles International Mean: 402 miles Mode: 1 mile Median: 79 miles Single Issue Mean: 200 miles Mode: 0 miles Median: 15 miles Left Wing n = 68 n = 33 n = 71 n = 262 Distances from Terrorist Residences to Incident Locations

Slide58: Residences to Activities

Slide59: 41% are 0-30 miles 20% are 31-90 miles 8% are 91-270 miles 19% are 271-810 miles 12% are 811-2430 miles Graphed Ranges Mean: 295 Median: 61 Mode: 0 Std.dev: 535 Distances from Terrorist Residences to Activities Locations n = 189

Slide60: Activities to Incidents

Slide61: Mean: 129 Median: 12 Mode: 0 Std.dev: 239 61% are 0-30 miles 10% are 31-90 miles 12% are 91-270 miles 13% are 271-810 miles 4% are 811-2430 miles Graphed Distances Distances from Terrorist Activities to Incident Locations n = 134

Breakdown of Ancillary Activity: Breakdown of Ancillary Activity

Breakdown of Planning and Preparatory Activities: Breakdown of Planning and Preparatory Activities

Residences to Incidents(data availability problems) : Residences to Incidents (data availability problems)

Slide65: For More Information: Brent L. Smith- Professor and Director Terrorism Research Center in Fulbright College Old Main 211, University of Arkansas, Fayetteville, Arkansas 72701 bls@uark.edu 479-575-3401 Kelly R. Damphousse- Associate Dean College of Arts & Sciences 633 Elm Ave, University of Oklahoma, Norman, Oklahoma 73019 kdamp@ou.edu 405-325-2529 Paxton Roberts- Research Associate Terrorism Research Center in Fulbright College Old Main 211, University of Arkansas, Fayetteville, AR, 72701 paxton@uark.edu 479-575-4521