Theodore Gordon a,⁎, Yair Sharan b,c,d, Elizabeth Florescu e,1
a Millennium Project, 1 Smilax Dr., Old Lyme, CT 06371, USAb Bar-Ilan University, Tel Aviv, Israelc TAM-C/FIRST Group, Jerusalem, Israeld Millennium Project in Israel Tel Aviv, Israele The Millennium Project Washington, DC, USA
a r t i c l e i n f o a b s t r a c t
Article history:
Received 28 May 2014Accepted 21 January 2015Available online xxxx
The recent terror events committed by individuals in the streets of Jerusalem, the CanadianParliament and in New York have raised attention to the phenomena of Lone Wolf terrorism.While there is no internationally-agreedupondefinition, generally a lonewolf (LW) is described asa single individual acting essentially alone who kills or injures people or significantly damagesinfrastructure to bring about political, religious, or ideological aims, to right perceived wrongs, togain recognition or in response to schizophrenic aberrations or other insane impulses. We call aLW with intent to use weapons of mass destruction resulting in a high number of casualties“SIMAD” standing for “Single Individual Massively Destructive.” A new kind of arms race isdeveloping. On the one hand,massively lethalweapons are becoming available to individualswhointend to use them, and on the other hand, methods of mass surveillance and for apprehendingsuch individuals are increasing as well. Will the methods of detection be adequate and timelyenough to avoid catastrophe? The outcomes of this race are important because weapons thatbecome available to LWs will have enormous destructive power, and interdiction in theiracquisitionmay involve new and repressive social constraints. Nomatterwhichway the scale tips,the world changes.To reach a broader understanding of this threat, the Israel Node of The Millennium Projectinitiated a Real-TimeDelphi study of prospects for LW terrorismwith the participation of some 60worldwide experts in the field of security and other related specialties. The participants generallyhad high confidence in their answers, but there were some sharp and important disagreements.We have early warning about this issue; we should use this time to better understand it, and takesteps to mitigate or avoid its threats.
We define lone wolf terrorists (LW) as single individualsacting essentially alone who kill or injure people or inflictsignificant damage on essential infrastructure at a single instantor over time, or plan to do so, in order to right perceived
wrongs, or accomplish political, social, religious, or ideologicalaims.
State sponsored terrorism offers many chances for interdic-tion and apprehension that are not possible for Lone Wolveswho intend to cause harm on their own. They may have beeninfluenced by organized extremism but they go about selecting,preparing for, and executing their adoptedmission on their owninitiative.Whydoes thismake interdiction harder? Since there isno “cell” involved, there is no chance for infiltration, no chancefor intercepting communications among co-conspirators, lesschance of using any detection strategies based on networks and
Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
connections, and no chance of partners in the prospective crimehaving second thoughts and alerting authorities. In other words,the interdiction modalities used to detect extremists' plots aremuch less effective against lone actors. That is why extremistorganizations urge decentralized and independent initiatives. Itfollows that a separate analysis is warranted.
Because an expanding LW threat seems plausible and verydangerous, we believe it is important to reach a broaderunderstanding of the potential for escalation in numbers andthe level of damage possible. Under the leadership of TheMillennium Project's Israeli Node, this objectivewas pursued ina Real-Time Delphi (RTD) study, involving experts from aroundthe world who represented the many skills required by thetopic. The design and findings of this RTD are presented in thisarticle and have been the subject of a presentation by theauthors at a NATO advanced research workshop conducted inNovember 2014 in Jerusalem, titled “Lone Actors—an EmergingSecurity Threat.”
Our definition is broader than that used by others; weinclude cyber-terror attacks, attacks against infrastructure,both successful and failed attempts, and attacks for apoliticalcauses (e.g. white supremacy, anti-abortion, animal rights,support or opposition to environmentalism, anti-Semitism).Cyber terror would be included because it may result inmassive social and economic disruptions and loss of life.Similarly, destruction or disruption of infrastructure mayaccomplish the objectives of some LW terrorists by causingmayhem and drawing attention to their activities and goals.Data on thwarted or failed attempts are sparse, but should beincluded if the success of interdiction strategies are to bemeasured.
Definitions of “lone wolves” vary among researchers in thisfield. For example, Spaaii2 defines lone wolf terrorists as thosewho “carry out attacks individually and independently fromestablished organizations.” Nesser,3 however, draws distinc-tions among leaderless resistance, solo terrorism, Lone Wolfterrorism, solo actors acting under control of a network, andsolo actors acting on their own but on behalf of organizedgroups. The European Research Consortium on TransnationalTerrorism defines LW terrorists as persons “who operateindividually, who do not belong to an organized terroristgroup or network, who act without the direct influence of aleader or hierarchy, whose tactics and methods are conceivedand directed by the individual without any direct outsidecommand or direction.”4 The definition used by Kushnerincludes lone wolves who might have been a “card carrying”member of some terrorist organization and even receivedfunds and training from them but “they take solitary actionwith only their conscience to guide them.”5
The lack of an agreed-to definition causes some difficulties.Is a sociopathic killer who acts alone a murderer, a LW, both, orneither? Some cases simply fall into a grey zone. A murderer isnot likely to be inclined to use a weapon of mass destruction;but even this is not a clear fulcrum: there have beenmurdererswho have brought down airliners in revenge and killed manypeople as an act of revenge or in an attempt to injure a singleperson.6 The line between criminal activities and terrorism isoften very thin.
Our definition excludes more than a few cases in whichpairs of people plan or perform acts of terrorism, such as thebombing of the Oklahoma U.S. Federal Building by TimothyMcVeigh who was assisted by Terry Nichols, the Tsarnaevbrothers who perpetrated the BostonMarathon bombings, andthe Palestinian cousins who killed three praying rabbis and afourth praying person in a synagogue on November 18, 2014.
A few of the most well-known lone wolf examples include:
Andres Breivik bombed government buildings and killedover 70 people in Oslo and on the island of Utyoa.
Ted Kaczynski, the Luddite Unabomber, killed 3 people andinjured 23 others with letter bombs.
Baruch Goldstein, an American-born Israeli physician, whokilled 29 and injured 125Muslims praying in the Cave of thePatriarchs in Hebron. 7
MohammedMerah, killed 7 some ofwhomwere children ata Jewish school8
Michael Zehaf-Bibeau opened fire at the Canadian NationalWarMemorial, killing one soldier and then running into theParliament building where he was killed by security andpolice forces.9
In the first of two shootings at Fort Hood, Major NidalHasan, killed 13 and injured 30 more.
In the Montreal École Polytechnique killing spree, MarcLépine killed 14 women students.10
In Newtown Connecticut, Adam Lanza killed 27 people,including an entire first grade class, at Sandy HookElementary School.
In the US Bruce Ivins, a researcher in the bio-defense lab inFort Detrick,mailed letters containing high quality antharaxresulting in 5 death and more then 60 casualties as well ashigh infrastructure contamination.
2 Spaaij, Ramon, “The Enigma of Lone Wolf Terrorism: An Assessment,”Studies in Conflict and Terrorism, Vol 33, Issue 9, 2010. http://www.tandfonline.com/doi/abs/10.1080/1057610X.2010.501426#preview.
3 Petter Nesser, “Single Actor Terrorism: Scope, Characteristics andExplanations, Perspectives on Terrorism,” Vol 6, No 6, 2012 http://www.terrorismanalysts.com/pt/index.php/pot/article/view/231/htmlhttp://www.terrorismanalysts.com/pt/index.php/pot/article/view/231/html.
4 Lone-Wolf terrorism, a case study by the European research consortiumTransnational Terrorism, Security and the Rule of Law (TTSRL); http://www.transnationalterrorism.eu/tekst/publications/Lone-Wolf%20Terrorism.pdf.
5 Kushner, Hatvey Encyclopedia of Terrorism, Sage Publications, 2003, http://militero.files.wordpress.com/2010/10/terrorism-encyclopedia-of.pdf.
6 In December 1987, David Burke shot the pilots of a Pacific Southwestpassenger jet in flight. The plane crashed and killed 43 people. The motive wasrevenge against the airline employee who had fired him and was a passenger.
7 Greenberg, Joel. “Israel destroys shrine to mosque gunman”, December 30,1999, New York Times.
8 Smith-Spark, Laura, “Who was French Gunman Mohammed Merah?”CNN, March 23, 2012. http://www.cnn.com/2012/03/21/world/europe/france-shooting-suspect-profile/.
9 CBC News, “Ottawa shooting: A day of chaos leaves soldier, gunman dead,”http://www.cbc.ca/news/politics/ottawa-shooting-a-day-of-chaos-leaves-soldier-gunman-dead-1.2808710.10 The Montreal massacre: http://www.theguardian.com/world/2012/dec/03/montreal-massacre-canadas-feminists-remember.
2 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
The list of incidents is much longer, of course. References tosome of the more prominent databases on terrorist and LWincidents—both accomplished and failed—appear in AppendixA. An analysis of these lists shows that:
In the period between 9/11/2001 and November 2014, LWshave accounted for 321 murders and injuries to about 900people. If the Oklahoma City and Boston bombings wereincluded, these numbers rise to 491 murdered and 1762injured. (Based on the Wikipedia list augmented by schoolshootings in U.S. and Canada.)
On prevalence, Spaaij concludes “lone wolf terrorismaccounts for 1.28% of all terrorist incidents from 1968 toMay 2007 in 15 countries”11 but Nesser in an independentreview concludes “Out of a total of 105 of planned, preparedand executed terrorist plots by jihadist actors between 1995and 2012 (based on the latest count), as much as 15 (14%)cases involved individual attackers”.12
The Heritage Foundation list of terrorist events in the U.S.that have been foiled or have failed between 9/11/2001 andApril 2013 includes some60 entries. Data onwhether or notan incident was perpetrated by a LW is not included in thisdatabase; however we classified the incidents and foundthat about twenty two of these foiled events were plannedor attempted by LWs (37%).13
Of the 22 LWs reported in theHeritage Foundation databaseof foiled and failed attempts, indictments of eight haveincluded reference to intent to use weapons of massdestruction (WMD).14
The bomb constructed and used by Timothy McVeigh todestroy the Federal Building in Oklahoma City couldwell beclassified as a WMD; it killed 168 people and injured 600.The pressure cooker bombs used in the Boston marathonterror killed 3 and injured 264. Since both examplesinvolved more than one person acting alone they falloutside of the LW definition used here but have beendesignated by some other researchers as LWs.15
The RAND database lists some weapons that have thepotential to become potential WMDs in terror incidents;these include anthrax biological agents, acids, ricin, Adamsite,cyanide, and chlorine gas, among others.16
The name we give to lone wolves who use or plan to useWMD is SIMAD. This termwas coined in a scenario prepared byTheMillennium Project in 2002, in a study on future science andtechnology management strategies. It stands for Single Individ-ual Massively Destructive and implies the threat posed by
psychopathic individuals who plan to do harm with weaponsthat are capable of killing and injuring large numbers of peopleor incapacitating infrastructure.17 In the 11 years that havepassed since the scenario was written, SIMAD has become aneven more chilling possibility. WMDs or their components maybe self-manufactured, obtained from military depots or labora-tories, or purchased on the black market.
While SIMAD is generally not currently recognized as asignificant problem, in the manner of futurists, it seemsprudent to study since it appears to be a plausible andthreatening possibility resulting from converging and reinforc-ing developments that include new technologies (such assynthetic biology), easier availability of WMD materials,publication of information necessary to construct them, and aproliferation of goals and incentives from the far left to the farright that could provide incentives for their use.
Stern warns that “As increasingly powerful weaponsbecome more and more available, lone wolves, who face fewpolitical constraints, will become more of a threat, whatevertheir primary motivation”.18 Similarly, Kushner notes: “Thereis little doubt that terrorists would use such weapons ifthey had them. Computer files captured from the Al Qaedagroup in Afghanistan in 2001 revealed that it was seeking toobtain chemical and biological weapons. The Aum Shinrikyocult in Japan prepared and used chemical and biologicalweapons in attacks on subway systems and other targets inthe mid-1980s”.19
While the possibility of weapons of increasing killing powergrows, so do the means of detection of insipient attacks. Somescientists argue that fMRI and other brain imaging tools couldhelp to identify potential terrorists. But if a threatening personcould be discovered before the fact, what would society doabout it? So far, all answers point toward restrictions of civilliberties.
2. Method: the RTD study
The well-known Delphi method was developed at RAND inthe late 1950s and 1960s to collect inputs from carefully chosenexperts in order to produce group judgments about topicsunder study such as forecasts or policies.20 21 As generallypracticed, the process involves a series of sequential question-naires, each building on the prior round. The participantsremain anonymous to one another and the answers of oneround are fed back in the next in order either to arrive atconsensus or understand reasons for disagreement. Usually,three or four rounds are required. While many studies havebeen accomplished using this approach, the process is generallyexpensive and time-consuming.
The use of Internet speeds the process by allowingparticipants to answer questionnaires on-line, while preserv-ing the basic principles of Delphi: anonymity of responses and
11 Op. cit. Spaaij.12 Op cit. Nesser.13 The listing of failed and thwarted attempts used in this analysis was preparedby the Heritage Group: http://www.heritage.org/research/reports/2013/07/60-terrorist-plots-since-911-continued-lessons-in-domestic-counterterrorism; theanalysis of this list was prepared by the authors.14 Ibid.15 Lone Wolf Attacks, Wikipedia; http://en.wikipedia.org/wiki/Lone_wolf_%28terrorism%29.16 http://smapp.rand.org/rwtid/search_form.php.
17 Glenn, J., Gordon, T., “2003 State of the Future”, The Millennium Project,Washington, DC. 2003.18 Stern, Jessica, “The Protean Enemy,” Foreign Affairs, July/August 2003.19 Op cit., Kushner, p. 422.20 For a detailed description of the process, see: Linstone, H and M. Turoff,“The Delphi Method,” http://is.njit.edu/pubs/delphibook/ (This is a freedownloadable book).21 Gordon, T. “The Delphi Method,” a chapter in Gordon TJ, Glenn JC. FuturesResearch Methodology – v3.0. Washington, DC: The Millennium Project, 2009.
3T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
feedback. In the 1970s Turoff experimented with computercommunications to link experts in on-line networks.22 Turoffalso experimented with an on-line social decision supportsystem that offered the possibility of participation by largenumbers of people.23 Researchers in Finland developed theeDelfoi system that has been applied to topics such as thefuture of education and identification of weak signals.24
In 2005, theUSDefense Advanced Research Projects Agency(DARPA) awarded a Small Business Innovation Research grantto Articulate Software Inc. to apply Delphi principles to tacticaldecision making to improve speed and efficiency.25 The opensource code that came out of this work was first listed onSourceForge and was further developed by Gordon at TheMillennium Project for use in its studies such as the State of theFuture Index, moral and ethical dilemmas of the future andfuture gender stereotypes.26 A further development of the RealTime Delphi process has been accomplished at TheMillenniumProject by incorporating it in its Global Futures InformationSystem (GFIS) that can be accessed at https://themp.org/.Several other researchers have developed their own Real-TimeDelphi systems, including Sabine Zipfinger at Johannes KeplerUniversity, Linz, Austria27 and Dr. Heiko von der Gracht at theCenter for Futures Studies and Knowledge Management, EBSBusiness School, Germany.28 The version developed by von derGracht and his colleagues includes innovations such as “an‘ease-of-use facilitator portal’, ‘consensus portal’, and a ‘graph-ical real-time feedback’, which reduces drop-out-rates andspeeds up the whole process.”29
Since Real-Time Delphi is conducted via the Internet,respondents can participate in a study whenever it is conve-nient to them and panels can be worldwide. A participant seesfeedback of numeric answers and reasons given for answers byothers after submitting their own, and may leave and return toa questionnaire without loosing previous answers. Some otheron-line Delphi systems, however, use the original multi-roundapproach, substituting communications on the Internet for theold paper and pencil approach.
Are the results of a Real-Time Delphi equivalent to theoriginal multi-round process? Gnatzy et al. say: “The researchfindings indicate that significant differences between the twoDelphi survey formats do not exist and final survey results arenot affected by changes in the survey procedure.”30
Careful selection of a panel is an important first step. For ourstudy of LW and SIMAD possibilities, an initial list of about 200potential contributors was compiled through a review ofliterature, recommendations of the 50 worldwide MillenniumProject Nodes, and “daisy chain” recommendations of peoplealready committed to participate. Ultimately, 57 people provid-ed answers to the questionnaire. Appendix B lists the names andother information about the respondents.
A small pilot group reviewed the first draft of thequestionnaire and offered comments and suggestions thatwere reviewed and incorporated in the operational question-naire design. The prospective participants were contactedpersonally or by email and invited to answer the online RTD.
The studywas performed over the period 10/20/2013 to 12/31/2013.
The RTD questionnaire was in the form of a matrix. The 17rows listed potential developments, variables, and policiespertinent to the evolving lone wolf problem; the columnsposed 2 sub-questions about the items that appeared on therows. One common sub-question was the respondents' confi-dence in their answer. As is usual in such studies, each cell hadseveral entries: a means of providing a response (such as radiobuttons), a summary of the group response, and a link to a newpage on which respondents were asked for the reasons behindtheir answers. Open-ended questions also permitted respon-dents to suggest other issues that might be added to the set thatappeared in the questionnaire. The questionnaire also included“mouse over” areas that expanded the definition of a row item,historical data where appropriate, and Internet links to author-itative papers and scientific literature for additional background.An excerpt from the questionnaire appears in Fig. 1.31
The instructions asked the respondents to be particularlyalert to problems of conflicting goals, for example, the possibletrade-offs between public security and individual liberties. Italso asked participants to avoid providing any information thatmight be considered classified.
The questionnaire included three types of questions:1) numeric, such as the expected year of occurrence of afuture event; 2) multiple-choice; and 3) text, which requiredwrite-in answers.
22 Turoff, M. Delphi Conferencing: Computer-Based Conferencing with Ano-nymity, Technological Forecasting and Social Change, 1972; 3: 159–204.23 Turoff M, Starr RH, Hee-Kyung C, Zheng L, Yuanqiong W. Social DecisionSupport Systems (SDSS), Proceedings of the 35th Hawaii InternationalConference on System Sciences, 2002.24 More information about eDelfoi is on http://www.edelfoi.fi.25 Gordon TJ, Pease A. RT Delphi: An efficient, “round-less” almost realtime Delphi method. Technological Forecasting and Social Change, 2006;73(4):321–333.26 Some other recent studies that used Real-Time Delphi include:
• Some Elements of the Next Global Economic System; How Might TheseNew Elements Affect the Human Condition over the Next 20 Years? 2009State of the Future, The Millennium Project.
• Future Elements of Art, Media, Entertainment; Millennium Project,Washington DC, 2008 Note: the questionnaire for this study is availableon line. Sign in at http://wwwrealtimedelphi.org. When a study code isrequested, type “arts” all lower case, no punctuation.
• White, Connie, Murray Turoff, Bartel Van de Walle, A Dynamic DelphiProcess Utilizing a Modified Thurstone Scaling Method: CollaborativeJudgment in Emergency Response, Proceedings of 4th InternationalConference on Information Systems for Crisis Response and Management,Delft, the Netherlands, May 13–16, 2007 Brussels University Press.
27 Sabine Zipfinger, Computer-aided Delphi: An experimental study ofcomparing round-based with real-time implementation of the method,Johannes Kepler University, Linz, Austria (2007); A review by T. Gordonappeared in Technological Forecasting and Social Change, 75, (2008) 160–164.28 Dr. Heiko von der Gracht, Tobias Gnatzy, Prof. Dr. Inga-Lena Darkow, Centerfor Futures Studies and Knowledge Management, EBS Business School,Germany and Theodore J. Gordon, Jerome Glenn, Millennium Project, “NewFrontiers inDelphi Research— Experienceswith Real TimeDelphi in Foresight,”a paper presented at WorldFuture 2011: Moving from Vision to Action, WorldFuture Society, Bethesda.29 Steffen Schuckmann, Marco Linz, Heiko von der Gracht, Inga-Lena Darkow,“Delphi Based Disruptive And Surprising Transformation Scenarios On TheFuture Of Aviation,” The 4th International Seville Conference on Future-Oriented Technology Analysis (FTA): 12 & 13 May 2011.
30 Gnatzy, T, Warth, J., von der Gracht, Darkow, I., “Validating an innovativereal-time Delphi approach— A methodological comparison between real-timeand conventional Delphi studies” Technological Forecasting and Social Change,Volume 78, Issue 9, November 2011, Pages 1681–1694.31 The full questionnaire can be accessed at http://realtimedelphi.org. Sign inthere and when prompted for a study code type in SIMAD (all caps, nopunctuation).
4 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
Fig. 1. Excerpt of the RTD questionnaire of the LW and SIMAD study.
Fig. 2. Quantitative questions results.
5T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
This first set of questions dealt with timing and escalation.Respondents who wanted to answer “never” were asked toenter the year 2100. The table in Fig. 2 presents the results tothe quantitative questions.
This is a remarkable result. The average answers fall on analmost straight line when plotted on a semi-log plot, as shownin Fig. 3.
There has been some discussion about whether a terroristtarget of 1,000,000 deaths is at all realistic. We note here thatthere have been several pandemics that have approached orreached this level. The third cholera pandemic of 1852–1860 issaid to have killed 1,000,000 people in Russia.32 The Spanish Flupandemic of 1918–1919 is estimated to have killed 50,000,000people worldwide.33 Furthermore, scientists have been able torecreate a copy of this deadly virus and test its infectiouscapability in laboratory animals.34 Floods and earthquakes alsokill high numbers of people, including the 1931 floods in Chinathat killed about 4 million.35
The numbers used to construct Fig. 3 are averages of theresponses of the RTD participants but average numbers do nottell the whole story. Despite the neat appearance of this graph,there was considerable disagreement among respondents.Fig. 4 shows how divided the estimates were.
The dichotomywas quite sharp: slightly over half the groupbelieved that an attack killing 100,000 people might occurbefore 2050 and the remainder of the group, by 2100 or never.
Further analysis of the data showed that there were twodistinct groups of respondents: those that identified “security”as their field (18 of 57) and those who self-identified as beingfrom other fields. The answers of these groups on timing of aSIMAD attack were quite different; the “security” set saw thethreat as occurring much later or not at all, as shown in Fig. 5.
As Fig. 5 shows, about 85% of the security experts answered“2100 or never” to the question of when a SIMAD might kill100,000 people in a single attack, but only less than 40% of thenon-security panel said “2100 or never.”
These results reflect only this groupof participants and, as inthe case of all Delphi studies, have little or no statisticalsignificance. They reflect only the opinions of these particularrespondents and another group might have very differentviews. Further, confirmation biasmight be atwork here. This is apsychological phenomenon that leads people to see data andevents through a lens that tends to confirm their biases.36 37 38
Respondents were invited to give reasons for their answers,and these were fed back to the group as well. One reason givenfor an early date of violence was:
Development of techniques and tools and knowledge… e.g.DIYbiotech, synthetic biology; decrease in personal restraint of
violent behavior globally, extensive training of violence from
Internet, former military, reactions of poor and oppressed used
as motivations, lack of spiritual growth in Y-generation and
millenials to lead to less self-restraint and hence gravitate to
violence
As for the late date, one respondent said he believed thatthere would not be a progression from Lone Wolf terrorism toSIMAD killings of massive numbers of people. He said:
Until now we didn't see a terror event with more then some
hundred casualties even not 9/11 (sic). High numbers can be
achieved by a LW only with a nuclear bomb which isn't
accessible to individuals. A bio epidemic will be stopped
quickly.
The next set of questions asked the respondents to choosean answer from a set of possibilities offered in the question-naire. In Fig. 6, below, each row presents a question and thecolumns show the alternative answers open to the respondentsand the number who chose each alternative. For example,Question 4 asked for a judgment about where a SIMAD attackmight first occur, and 25 of the 45 respondents chose “NorthAmerica.” In discussing geographic locations, a respondent said:
There are strong recent historical reasons for postulating such
an attack on a North American (combined US–Canada) target,
given the interconnectivity of the grids, but equally there are
Fig. 3. Trend of casualties of a SIMAD attack.
32 Third cholera pandemic (1852–1960) https://en.wikipedia.org/wiki/Third_cholera_pandemic_(1852–60).33 Spanish Flu, Science Daily, http://www.sciencedaily.com/articles/s/spanish_flu.htm.34 Conner, Steve, “American scientists controversially recreate deadly SpanishFlu virus,” The Independent, 11 June, 2014.35 The Great Flood of 1931 at Gaoyou, September 25, 2007, http://aboutgaoyou.com/history/floods/the_floods.aspx.
Fig. 4. Estimates of when a 100,000 casualties SIMAD attack might occur.
36 Totlock, P.E., “Theory-Driven Reasoning about Possible Pasts nd ProbableFutures: AreWe Prisoners ofOur Preconceptions?”American Journal of PoliticalScience, 43, 335–366, 1999.37 Kahneman, Daniel, “Thinking Fast and Slow,” New York, 2011.38 Arkes, H., “Overconfidence in Judgmental Forecasting,” a chapter inPrinciples of Forecasting, J. Scott Armstrong (ed), University of PennsylvaniaWharton School, Springer, 2001.
6 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
strong reasons to believe such attacks could occur elsewhere,
particularly in Russia (because of Islamist/khanate region
disaffection), and also in the PRC.
The next question dealt with motive. The respondents felt,overwhelmingly that two motives were the most important:religion and righting perceived wrongs. Agreement was alsostrong on potential targets: they were population at large andspecific population segments. Some respondents introducedother possible targets:
I have answered population at large, but agriculture runs a
close second, particularly if we imagine a binary anti-crop
weapon, which can hold harvests at ransom. Terrorists hold
the antidote to the poison they have administered to a crop and
they extort money or demand actions to allow the harvest.
And another said:
Where government targets are not available, public symbols
and social groups will substitute.
They largely agreed that there would be an effort to findpotential LWs before they could act (as is now the case) andthat the techniques used in finding them were likely to be on-line monitoring of purchases, tracking communications, andthrough tips from 3rd parties (as is now also the case). Futuretechnologies that might be used for early identification ofpotential LWs (genetics and brain research) got few responses(although progress in these fields is promising39 40 41 42).
Rohn and Erez, inwriting about prospects for agro terrorismadd an interesting dimension to the anticipatory search forterrorists. They suggest that there are “styles” in terror that canhelp assign priorities in on-line searches for incipient terror andterrorists. For example, blowing up buildings was popular inthe late 20th century, then beheadings “became trendy,” andthey say: “The antibiotics-resistant salmonella found in Europeduring the summer of 2011 along with analysis of currentaffairs by CATRC (The Counter Agro Terrorism Research Center,Israel) suggests that the next trend in Agro-Terrorismmight aswell be the disruption of a central hub in the food chain byagro-terror means rather than brute force.”43
Can any strategy avoid potential SIMAD attacks? Threequarters of the respondents agreed that effort would be madeto detect and divert SIMAD activities using profiling (someprofile elements might include: fascination with mass killings,playing violent videogames, difficulty in social communica-tions, possibility of being bullied as a child, proficiency withfirearms, seizures, lack of emotional connections, and obsessedabout battles, war, and destruction), eavesdropping and socialmedia. One respondent also suggested searching for a geneticcomponent that could predispose certain people toward thiskind of violent behavior, and thus raising the possibility ofgenetic screening.
Given a number of possible choices of the fields fromwhich lonewolf weaponswould be derived, the respondentsfavored biotech and synthetic biology, and nuclear physicsover other possibilities. There were also some other sugges-tions including:
Invasive species as a bioweapon VERY easy to make, distribute,
and have severe impact on nations, public health, ecosystems,
global food production, and commodity markets.
Nanotech produces agents that are penetrative, saturational,
invisible, destructive, and dispersible. Potentially, biggest bang
for the buck, even more so than biological agents though nano
can be combined with bio.
Cyber terrorism is the most likely vehicle for successful,
sustained attack by lone wolf
I think they might go for very simple weapons that have mass
killing implications, such as very effective poisons, distributed
in novel ways.
Some 60% of respondents thought that in the long run “softapproaches” such as public awareness campaigns and educa-tion reform might be effective in reducing the threat ofterrorism, within limits:
While soft approaches might work fairly well in most cases,
inevitably they will not be 100% effective and therefore the
threat will exist still… such approaches …. will result, over
time, in a smaller pool of people willing to engage in lone wolf
terrorism. But unless the global system undergoes dramatic
change, therewill always be those aggrieved or twisted enough
to pursue violence, even if only for its own sake.
Fig. 5. Estimates of LW casualties by groups of respondents.
39 Telling Truth From Lie In Individual Subjects With Fast Event-Related fMRI.Langleben DD, Loughead JW, Bilker WB, Ruparel K, Childress AR, Busch SI, GurRC. Hum Brain Mapp. 2005 Dec;26(4):262–72. Medline: http://www.ncbi.nlm.nih.gov/pubmed?term=Langleben%20DD%5BAuthor%5D&cauthor=true&cauthor_uid=16161128.40 Personality correlates of electroencephalographic patterns: Rorschachfindings. Rabinovitch, M. Sam; Kennard, Margaret A.; Fister, W. P. CanadianJournal of Psychology/Revue canadienne de psychologie, Vol 9(1), Mar 1955,29–41.41 Jennifer Granick, The Lie Behind Lie Detectors, Wired 2006: http://archive.wired.com/politics/law/commentary/circuitcourt/2006/03/70411?currentPage=all.42 Terrie Moffitt, Avshalom Caspi, Joseph McClay, Jonathan Mill, Judy Martin,Ian W. Craig, Alan Taylor, Richie Poulton, “Role of Genotype in the Cycle ofViolence in Maltreated Children,” Science 2 August 2002: Vol. 297 no.5582 pp. 851–854: http://www.ncbi.nlm.nih.gov/pubmed/12161658.43 Rohn, Eli, and Erez, Gil, “Fighting Agro Terrorism: A Framework forIntention Detection Using Overt Electronic Data Sources,” Intention DetectionUsing Overt Electronic Data Sources, Proceedings of the 9th InternationalISCRAM Conference. Vancouver, Canada, 2012 http://www.iscramlive.org/ISCRAM2012/proceedings/261.pdf.
7T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
There was an important split in answers to the questionabout whether scientific and technological papers containinginformation that might be useful to terrorists should be
withheld. More than half the panel felt that such papers shouldnot be withheld even if they contained information of potentialuse to the terrorists. Why? As one respondent said:
Multiple Choice Questions and Answers
4. Where do you think an
actual lone wolf terrorist
attack of the sort described
in Question 3 might first
occur?
North
America
Middle
EastEurope
Asia (ex
China)Africa
South
AmChina
25 9 8 2 1 0 0
5. However distorted the
thinking, what do you thin
is the single most importan
motivation that will fuel
lone wolf attacks over the
next decade?
ReligionRight
WrongsInsanity Govmt History
Mone
yOther
19 17 3 2 2 1 1
6. What do you think the
primary target of lone wol
attacks will be over the
next decade?
Population
at large
Population
SegmentsInfrastruct Officials Other
Agri/
Food
21 16 5 4 1 0
7. Do you think serious
attempts to search for lone
wolf terrorists who are
capable of carrying out an
attack using a weapon of
mass destruction will be
made before such an attack
occurs?
Yes No
33 13
8. What technology is
likely to be most effective
for the detection of people
with evil intentions?
Consider fields such as
psychology, brain imaging
observation of unusual
behavior, etc.)
Monitor
Purchases
Communic
ations3rd Parties
Psychologic
ScreeningGenetic
Brain
Resea
rch
Other
18 14 6 4 3 0 0
10. Do you believe that
scientific and technologica
papers and other
publications that contain
information potentially
useful to terrorists should
be controlled or withheld?
Yes No
18 25
11. Assume that some lon
wolf terrorists choose to
use massively destructive
or disruptive weapons;
from what fields might
these weapons come?
Bio TechComputers/
Comm
Agriculture
and foodOther
Power
Generat
ion and
Trans
Nucle
ar
Physi
cs
Nano
Tech
27 6 4 4 2 2 2
12. Are "soft" approaches
such as education reform o
public awareness
campaigns likely to be
effective in dealing with
the lone wolf threat in the
long term?
Yes No
28 17
13. Are actions which
intrude on privacy of
people or otherwise
compromise their civil
rights justifiable in view o
the threats?
Yes No
24 22
14. If people are identified
as potential lone wolves of
the sort we have already
seen currently, how do you
think society should deal
with them?
Incarcer-
ation
Monitoring
Watch List
Rehab
Psycho-
therapy
Due Process
Medical
Treatme
nt
Other
16 11 9 5 2 2
Fig. 6.Multiple choice questions and answers.
8 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
Because the very nature of scientific research is in its
publication
Similarly the panel was split on the question of whether ornot intrusion into privacywas justified in an effort to limit LoneWolf terrorism.
Yes, because of the potential for mass casualties.
No. The resolution “Right to privacy in the digital age” adopted
by the UN General Assembly in December 2013 is rightly
calling on all countries to take measures to end activities such
as electronic surveillance, interception of digital communica-
tions and collection of personal data, which violate the
fundamental “tenet of a democratic society.”
One respondent asked the poignant question: who will
control the controllers? This debate has begun and judging fromthe responses of this group will continue.
How should society deal with people who are identified aspossible LWs? Answers: incarceration, watch lists, and rehabpsychotherapy.
The last four questions posed to the respondents wereopen-ended:
14. If people are identified as potential lone wolves of the sortwe have already seen currently, how do you think societyshould deal with them?
15. What if they have been engaged in building weapons thatcould produce mass destruction or disruption?
16. What steps, if any, do you think should be taken tominimize lone wolf threats? When?
17. What steps, if any, do you think should be taken tominimize threats thatmay achieve destruction on the scaledepicted in Question 2? When?
Many suggestions were made in response to these open-ended questions about steps that should be taken tominimize LW threats. Many responded with support formonitoring, control, education, and intelligence gathering.In addition, consider these suggestions for mitigating LWthreats:
When new-technology weapons are developed, (there is an)
obligation to also develop the antidote for them or if not
possible, moratorium on the use of the respective technology.
Ban onweaponizable bio/nano technologies. Ban possession of
weapons by citizens.
Long term: education and reduction of poverty; higher level
public discourse. These are long-term, however, and probabi-
listic. More immediately: sensors at key transit points and
supply sources; focusedmonitoring ofmovement of potentially
destructive materials—muchmore effective than data sweeps
and tracking every blip in cyber-space.
Minimizing this threat is a long time continuous effort
for the national and international authorities that imply
and hard measures should be combined according to the
specific situations. There is no magic formula to stop this
phenomenon.
Only a single person of the 44 that responded felt that ALL(100%) of SIMAD attacks could be averted.
Suppose that methods were invented in the future toidentify people who might become LW. How should societydeal with them? Some of the respondents warned that fearof large scale violence could lead society down the path tointrusive, authoritarian, and lawless government. Otherssuggested that due process within the current legal systemwould be adequate. Two thirds suggested incarceration,removal from society, monitoring, observation or placingon a watch list. The remainder suggested rehabilitation,counseling, psychotherapy, and medical treatment. Onerespondent summarized:
We cannot lock people up for what they might be thinking
absent proof of intent to commit. Monitoring is a balanced
approach…. If they have not broken any laws, observation. If
they have or are about to commit an attack, interception and
due legal process.
But what if they have been engaged in building weaponscapable of mass destruction or disruption? How then shouldsociety treat potential terrorists? Suggestions by the respon-dents fell into the same categories as the previously noted;however some new suggestions were made:
Consider the possibility of special courts/tribunals in which the
judges include psychiatrists to examine and prescribe for the
suspect
Not enough or not soon enough will always haunt us.
Totalitarian societies seldom have this dilemma.
A final question asked what steps might be taken tominimize threats of large-scale SIMAD-like destruction if suchthreats were to become even more palpable. Some of theinteresting ideas included:
It is possible that a genetic component exists which induces
aberrant behavior. If that is so, genetic manipulation might
also be considered. Of course this brings important civil liberty
questions to the surface.
Several participants felt that media has a large role to play.
Minimizing media hysteria is the key to removing the
appeal of all forms of terrorist behavior, including lone
wolf actions, but can media hysteria be reduced without
impinging on fundamental freedoms? Or can media respon-
sibility be developed?
Media should be full time involved in sending subliminal
messages, so people can reflect about their own behaviors.
The media should not “heroize” those that commit mass
destruction or aggression of any kind.
Perhaps sometime in the future itwill be possible to identifyincipient LWs from a random group of people. One respondentsaid:
The state of Connecticut's report on the Shootings at Sandy
Hook Elementary School list these observations about the
shooter; can any of these, or the assemblage of all of them, form
9T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
the basis for an early warning system designed to identify
potential SIMAD's: fascination with mass killings, playing
violent videogames, difficulty in social communications, possi-
bility of being bullied as a child, proficiency with firearms,
seizures, lack of emotional connections, and obsessed about
battles, war, and destruction. Similar lists might be derived from
studies of othermassmurderers. Also consider that theremay be
a genetic component so that DNA analysis may play a part- or
analysis of behavior of parents.
Bakker and de Graff point out: “… it is particularlydifficult to differentiate between those lone operator extrem-ists who intend to commit attacks and those who simplyexpress radical beliefs or issue hollow threats (hoaxes) …
While most terrorists are radical but not all radicals areterrorists.”44
Were such screening to be attempted, many legal and socialissues would arise. Would the stigmatized individuals bedenied due process and certain freedoms?
4. Discussion: how LW terrorism can morph into SIMAD
4.1. The scenario
The pieces of a SIMAD scenario seem to be falling intoplace. First, consider that remotely controlled drones areproliferating. Delivery mechanisms for bio-weapons areeven simpler. Aerosol forms of an infectious virus orbacterium can be easily dispersed, or even delivered inthe form of a sneeze by an infected person. One participantin the RTD said:
Availability of biological weapons even made from scratch
exists presently; use of invasive species as biological weapons
exists presently, nuclear weapons and radioactive materials to
make radiological bomb-presently exists; chemicals to make
chemical weapons presently exist, Internet information to
make EMP pulse weapons presently exists; capability to make
such weapons airborne with balloons or put into confined
population areas such as sports arenas, malls, or public
political events presently exists.
In this response “invasive species”might include the use ofmicroorganism or insects as biological agents and as pointedout by Wilson et al.; Cuba in the 1960–1990s accused the U.S.of deliberately using agents such as these to disrupt theireconomy.45 Wilson and his colleagues, drawing on priorliterature, also provide a list of selection criteria fordangerous antilivestock and antipoultry biologic agentsand 22 agents that provide the greatest threat. There isvery little overlap between this list and a list of agents thatare threats to humans.
Synthetic biology is a new and growing professionalfield that has attracted a set of hobbyists who practicethe new science as advanced amateurs. They often try todevelop more efficient bio-fuels, improve photosynthesisefficiency, and develop hobbyist instruments46 or simplydemonstrate their capabilities by producing bio-exoticasuch as glowing plants. They buy biological laboratoriesand biological parts (e.g. promoters, proteins, DNA,RNA, coding sequences, plasmids, terminators and othergenetic molecular components) from open sources on theInternet.47
A SIMAD could use DIY synthetic biology to develop aWMD capable of killing large numbers of people or affectingplant or animal agriculture. Couple easy availability ofmaterials for weapons of increasingly destructive power,easy means of delivery, and the high intensity focus of themedia, and the progression from LW to SIMAD seems quiteplausible.
On the other side of this coin, some of the technologiesuseful to SIMADs also hint at means for their deterrence. Bio-hacking can lead to new hybrid plants and organisms thatare designed to detect atmospheric pathogens48 or earlydetection of contaminated food and water supplies.49 50 51
The IBM computer Watson is now being used in the analysisof large scale data bases. Perhaps its inferential powercould be harnessed to search the huge repositories of videorecords using facial recognition to identify specific people atgiven times and places.52 Not only recognition of specificpeople, but recognition of emotions 53 could be importantin interdiction of SIMADs; the Future Attribute ScreeningTechnology (FAST) system being designed by the US Officeof Homeland Security is intended to search for individualswith mal-intent, as though mal-intent were a disease and
44 op cit. Bakker and de Graff.45 Wilson, Terrance M., Linda Logan-Henfrey, Richard Weller, and BarryKellman, “Agroterrorism, Biological Crimes, and Biological Warfare TargetingAnimal Agriculture,” Emerging Diseases of Animals, Edited by C. Brown and C.Bolin 2000, ASM Press, Washington, DC.
46 Heidi Ledford, “Garage biotech, Life Hackers,” Nature, Published online 6October 2010 | Nature 467, 650–652 (2010) | http://dx.doi.org/10.1038/467650a.47 See: Registry of Standard Biolog Giraldo, J. P., et al., “Plant nanobionicsapproach to augment photosynthesis and biochemical sensing,”: NatureMaterials, March 9, 2013; http://www.nature.com/nmat/journal/v13/n4/full/nmat3890.htm.48 Heidi Ledford, “Garage biotech, Life Hackers,” Nature, Published online 6October 2010 | Nature 467, 650–652 (2010) | http://dx.doi.org/10.1038/467650a.49
“Mercury contamination in water can be detected with a mobile phone,”Science Daily, February 6, 2013; http://www.sciencedaily.com/releases/2013/02/130206111027.htm.50 Volinsky, Alona, “NewTest ToDetect Food ContaminationWithinMinutes.”Israeli Innovation News, Novemebr 17, 2011; http://nocamels.com/2011/11/new-test-to-detect-food-contamination-within-minutes/.51 Amit Singh, Somayyeh Poshtiban, Stephane Evoy, “Recent Advances inBacteriophage Based Biosensors for Food-Borne Pathogen Detection,” Sensors2013, 13, 1763–1786.52 IBM forms new Watson group for cloud-delivered cognitive innovations,with $1 billion investment, January 12, 2014, http://www.kurzweilai.net/ibm-forms-new-watson-group-for-cloud-delivered-cognitive-innovations-with-1-billion-investment?utm_source=KurzweilAI+Weekly+Newsletter&utm_campaign=dd51a10e0f-UA-946742-1&utm_medium=email&utm_term=0_147a5a48c1-dd51a10e0f-282100129.53 Scharr, Jillian, “Facial-Recognition Tech Can Read Your Emotions,”livescience, January, 2014; http://www.livescience.com/42975-facial-recognition-tech-reads-emotions.html.
10 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
the factors involved in the screen were symptoms of thatdisease.54
Third party reporting is also important. A 2012 study,sampled over 1500 adults in the United States to determinetheir attitudes about reporting certain kinds of behavior to thepolice. Some 90% of participants said they would be “verylikely” or “somewhat likely” to call the police if they heard aperson talking about planting explosives and 50% if theysaw someone reading material from a terrorist group.55 Lawand professional ethical standards already require somereporting; in the United States; for example, psychiatrists,physicians, psychologists, lawyers, and counselors are expectedto report certain potentially threateningbehavior to authorities.Training people (particularly teachers, parents, doctors, socialworkers, etc.) on symptoms of thepotential for violent behavioris a good idea that was suggested by a referee who reviewedthis paper.
Work is also progressing on the use of functional MRI toidentify brain anomalies associated with mal-intent,56 andgenetic analysis to search for genetic markers of potentiallyviolent persons. One team found that a combination of anabused childhood and low activity of promoter levels for themonoamine oxidase-A (MAOA) gene resulted in high propen-sity for anti-social behavior. The lead researcher reported,“These findings provide the strongest evidence to datesuggesting that the MAOA gene influences vulnerability toenvironmental stress, and that this biological process can beinitiated early in life.”57
About half of the respondents to our RTD study said theybelieved that a SIMAD scenario could be avoided. Seventy fivepercent of the RTD respondents believed that serious attemptsto search for SIMAD terrorists would be made before an attackoccurs. Two respondents pointed out current efforts at pre-detection:
FBI is already using trip wires (sales/distributors of hazardous
chemicals and precursor chemicals) in place to provide
information on suspicious sales.
Precisely what NSA is currently accused of—eavesdropping
(spying), on *everyone* is what it will take. Despite this
activity running contrary to the American vision/expectation,
it is the most effective means of learning what someone is
thinking about, planning to do.
The significant question is whether or not such techniquescan be effective in identifying potentially dangerous peoplebefore a SIMAD catastrophe occurs. Only 25% of the RTDrespondents thought that 80–100% of SIMAD attacks could beaverted. And only one respondent said 100% could be thwarted.Fig. 7 shows the respondents' judgment and lack of agreementabout averting SIMAD attacks.
4.2. The cyber dimension
The Internet can be used by extremist organizationsfor recruiting, education, communications, and fund raising.Gabriel Weimann notes 58 that similar to wolves—which neverhunt alone, but in packs—most LW terrorists “connect,communicate, and share information, know-how, and guid-ance” using the Internet. LW terrorists can also attack throughthe Internet. As hackers, they can act alone without attractingmuch attention. Furthermore they can execute their planswithout threat of physical harm to themselves. A LW couldcause large-scale destruction, for example, by causing vitalservices such the electrical transmission grid or the interna-tional flow of funds, to crash. Messages on the Internet canglorify and inspire copycats.
The U.S. Department Of Homeland Security reported: “In1998 there were…12 terrorist relatedwebsites active. By 2003there were approximately 2630 sites, and by January 2009 atotal of 6940 active terrorist-related websites.”59
Among the relevant comments from RTD were:
Cyber terrorism is the most likely vehicle for successful,
sustained attack by a lone wolf.
If cyber terror is included (it ought to be) then the answer (to
when attacks will begin) is now.
Consider including Cyber Terrorism as a WMD. Cyber terror
has been defined as: the intimidation of civilian enterprise
through the use of high technology to bring about political,
religious, or ideological aims, actions that result in disabling or
54 Future Attribute Screening Technology Project, US Department Of Home-land Security, December, 2008: https://www.dhs.gov/xlibrary/assets/privacy/privacy_pia_st_fast.pdf.55 Testimony of William Braniff, Executive Director National Consortium forthe Study of Terrorism and Responses to Terrorism (START), to the HouseCommittee on Homeland Security Subcommittee on Oversight and Manage-ment Efficiency “Why Can't DHS Better Communicate with the AmericanPeople?”May 21, 2013. http://docs.house.gov/meetings/HM/HM09/20130614/100875/HHRG-113-HM09-Wstate-BraniffB-20130614.pdf.56 Telling Truth From Lie In Individual Subjects With Fast Event-RelatedfMRI. Langleben DD, Loughead JW, Bilker WB, Ruparel K, Childress AR, BuschSI, Gur RC.Hum Brain Mapp. 2005 Dec;26(4):262–72. Medline: http://www.ncbi.nlm.nih.gov/pubmed?term=Langleben%20DD%5BAuthor%5D&cauthor=true&cauthor_uid=16161128.57 Nurturing Nature, The Economist, August 1, 2002. http://www.economist.com/node/1259045.
Fig. 7. Distribution of responses to averting SIMAD attacks.
58 Gabriel Weimann. Virtual Packs of Lone Wolves https://medium.com/p/17b12f8c455a.59 US Department of Homeland Security, Washington DC, citing a presenta-tion given byDr. GabrielWeimann at the Youth Recruitment and RadicalizationRoundtable, March 19, 2009.
11T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
deleting critical infrastructure data or information and/or
resulting in massive loss of life.
The Internet also can be used to deter or anticipate terroristattacks in several ways: by tracing email and social mediamessages back to their origins, by simply reading what isposted on extremists web sites (LW often post plans anddescribe achievements), as an offensive weapon as in theStuxnet case,60 by identifying who is communicating withknown terrorists, by tracing purchases of critical or unusualmaterials on public sites aswell as on Tor, “darknet”, or throughthe use of software programs that scan all pertinent databasesand match factors—behavioral and otherwise—that point to atroubled future.
The cyber attack on Sony Pictures (thought to have beeninitiated by North Korea) only begins to illustrate howdangerous cyber terror can be.61
4.3. Social impacts
The comments of participants in the RTD indicate thenecessity of developing security and resilient strategies toprotect against LW. Clear international legal frameworks areneeded for counterterrorism measures of all types—fromdetection and scrutiny techniques, to control of potentialweaponizable technologies. Using intrusive surveillance with-out unambiguous legal authority and public transparencycould undermine the legitimacy of the countermeasuresthemselves.
Regulatory agencies have a propensity to respond to crisesonce they have occurred; acting proactively is difficult. Yet 75%of the RTD respondents believed that serious attempts to searchfor SIMAD terrorists would be made before an attack occurs.
Social media, such as YouTube, Twitter, Facebook andothers have become new “herding” and incubator spaces fornew ideas, including security-threatening ones. Should therebe constraints on online magazines like Inspire, or on Facebookpages that preach radicalization and teach how to buildexplosive weapons? Respondents were split sharply on thisissue.
Genetic analysis or some future version of fMRI mightbe used to scan for brain abnormalities that could beassociated with aggressive or violent behavior. However,there are as yet no legal frameworks for DNA collection andbrain scanning, psychological testing, or automated reviewsof school or health records. Biometric data—digital recordingof fingerprints, facial characteristics, iris scans, and thelike—might be used to identify people with characteristicsthought to be related to terrorism (since 2003, the U.S.Department of Homeland Security has spent over $133million, and the Department of Defense budgeted $3.5billion over 2007–2015)62; but suppose the databases were
hacked. Hacked or not,what happens to the people stigmatizedby this surveillance?
Similarly, synthetic biology, virus modification, and othersuch technologies raise new environmental, health, andsecurity concerns. The World Health Organization is stilldebating whether the United States and Russia shouldeliminate their strains of the live smallpox virus, since it is sodangerous. Through synthetic biology, essentially any viruscould be modified for use in biological terrorism. “Syntheticbiology adds a new wrinkle …,” warns Jimmy Kolker, U.S.Health and Human Services Assistant Secretary for GlobalAffairs; “We now aren't as sure that our countermeasures aregoing to be as effective as we'd thought even five years ago.”63
5. Possible overconfidence
The first 13 questions in the RTD called for the respondentsalso to provide judgments about their confidence in theiranswers. Their responses indicated a remarkably high level ofconfidence, as shown in Fig. 8.
This is remarkable since the events that were being judgedare so intrinsically uncertain and in many cases unknowable.As the psychologist Daniel Kahneman says, “high subjectiveconfidence is not to be trusted as an indicator of accuracy(low confidence could be more informative.)”64 Further-more, the literature on decision-making and judgmentmakes the point that experts are often overconfident abouttheir intuitions.65 66
6. Conclusions
1. Some places in the world are likely to be targeted more than
others. (See answers to question 4) The RTD paneloverwhelmingly selected North America. One panelist,however, felt that Asia was more likely to be the initialtarget. His reasoning was:
[The SIMAD attack would likely be in a place] with large
population and geographic base, conflict is significant. Access
to means could be high. Visibility can be moderate to high.
Security systems to prevent may be porous. Belief systems are
strong and polarizing…
2. Data should be collected to help identify attractive LW and
SIMAD targets so that those places can be given priority in
planning security measures.
(Respondent comment in response to question 4.) Apanelist suggested that such criteria might be usedto form a prospective “hit list.” Some panelists madethe point that targets could be unconventional; forexample, they mentioned agriculture, and specific popula-tion segments.
60 Kushner, David “The Real Story of Stuxnet: How Kaspersky Lab trackeddown themalware that stymied Iran's nuclear-fuel enrichment program,” IEEESpectrum 26 Feb 2013 http://spectrum.ieee.org/telecom/security/the-real-story-of-stuxnet.61 http://www.nytimes.com/2014/12/18/world/asia/us-links-north-korea-to-sony-hacking.html?_r=0.62 The Boring and Exciting World of Biometrics http://www.pbs.org/wgbh/nova/next/tech/biometrics-and-the-future-of-identification/.
63 Some Experts Urge Eradication of U.S., Russian Smallpox Stocks http://www.nti.org/gsn/article/foreign-experts-urge-eradication-us-russian-smallpox-stocks/?mgs1=99a7fJOxUZ.64 Kahneman op. cit.65 See, for example: Angner, Erik, “Economists as Experts: Overconfidence intheory and practice,” http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.118.9725&rep=rep1&type=pdf.66 Kahneman, Daniel, Paul Slovic, Amos Tversky. Judgment under uncertainty:
Heuristics and biases. Cambridge University Press. pp. 306–334. ISBN 978-0-521-28414-1.
12 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
3. Identify means for restricting media reports of LW and SIMAD
episodes, if possible, without undue impact on press and
speech freedom.
The media must communicate the threat, but theyshould not offer prospective terrorists public theaterthat would create copycats and incite others to thesame behavior. (See comments of several respondentson page 20.)
4. Study and implement policies that inhibit availability of
potentially dangerous scientific data and information.
(See responses to question 10.) The panel split on thestrategy of withholding potentially damaging scientif-ic and technological papers, even if they containinformation of potential use to terrorists because ofthe need for freedom of scientific publication. Howcan this information be made available for scientificpurposes without amplifying the killing potential ofLWs?
5. The size of a potential SIMAD attack may increase exponen-
tially with time.
The panelists were asked for their judgments about thelikely timing and size of an initial SIMAD attack. Plottingtheir average answers as a semi log graph results in astraight line; this implies a constant rate of growth in thenumber of people killed in a SIMAD attack of 12% peryear. However there was considerable disagreementand widespread differences among the responses. (SeeFig. 3)
6. The timing of a SIMAD attack is very uncertain and according
to RTD responses may never happen or may happen sooner
rather than later.
When respondents were asked when, if ever, 100,000deaths might occur due to a SIMAD attack, the distribu-tion of judgments was bimodal. One interpretation isthat if there has not been a SIMAD episode by 2050, itwould be delayed for decades or may never happen.Almost all of the people who placed the timing of theattack as “Between 2075 and never” self identifiedthemselves as experts in “Security.” Is this confirmationbias? (See Figs. 4 and 5.)
7. The computer can be a weapon, even a weapon of massdestruction of systems that are required for society tofunction. The RTD participants mentioned manypossible targets. (See discussion of cyber-terror inSection 4.2)
8. Weapons used by LWs may be unexpected.The panelistsmentioned several—some low tech—that might haveweapons implications: new effective poisons distributedin novel ways, invasive species as a bioweapon, anddispersible nanotech agents.
9. Three categories of measures designed to identify or influence
persons who could become SIMADs were suggested.“Softapproaches” (e.g. public awareness campaigns, em-phasis on family, social justice, and ethics, and educa-tional reform). “Hard approaches” (e.g., geneticscreening, fMRI brain scanning, psychological condi-tioning, and “Vigilant monitoring” of people thought tobe predisposed to violence). (See answers to question12.)
10. New technologies could help detect propensity to violence and
should be vigorously pursued including psychological ex-amination, brain imaging, profiling, and perhaps genetictesting. (See discussion in Section 4.3.)
11. Attempts to identify SIMADs before the fact raise
questions about privacy, civil liberties, and freedom.
The RTD panel split on the question of whether or notintrusion into privacy was justified. (See discussionSection 4.3.)
12. Religious leaders have an important role to play.The panelidentified religious incentives as the single mostimportant motivation behind future LW attacks. Whenit becomes clear that LW can morph to SIMAD andmany thousands could be killed in a single event,will leaders of religious movements still encourageLW attacks in certain circumstances? (See answers toquestion 5.)
13. Addressing this threat requires a long time continuous effort
by national and international authorities including dataand intelligence sharing. Cooperative efforts are ofcourse underway, but the new perception of SIMADimplies that these efforts must expand and continue fordecades.
14. Several methodological issues were apparent and deserve
further study; these include understanding the meaning ofhigh confidence when the answers were obviously so veryuncertain. (See Fig. 8.)
15. Many important questions raised in this study have not been
answered. The panel of experts came to no consensus anddisagreed about issues such as:
• The effectiveness of means of detection of LWs andSIMADs
• The possibility that LW terrorism would expand intoSIMAD, and if it were to occur, its timing and scope
• The appropriate trade-off between civil liberties andimproved security
• The right to publish scientific papers that containpotentially threatening “how to do it” weaponinformation
• Means of coping with people who are suspected ofhaving SIMAD potential
These important issues deserve much further attention.NB. Does our work present new possibilities for future
lone wolves? We took this possibility into considerationwhen designing the Real Time Delphi and in writing thispaper. In the RTD, the instructions to the respondents
Fig. 8. Confidence of respondents in their answers.
13T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
specifically asked that they exclude from their answersany information that might be considered classified. Inaddition we asked about the fields from which Lone Wolfweapons might be derived, rather than descriptions of thespecific technologies themselves. While the manuscriptmentions many weapons and detection techniques, noneseem revolutionary and able through their mention here tolend power to those with mal-intent.
Acknowledgments
As in any study of this sort, there were many contributors.First and foremost among these were the expert participantsto the Real-Time Delphi study who freely contributed theirtime and thoughts in response to tough questions aboutdark futures. They helped us understand the importance anddifficulty of identifying proto SIMADs and preventing theirattacks without creating a totalitarian nightmare. We alsoappreciate the comments of reviewers who helped us bymaking many suggestions about the material. This work wasself-funded, the participants received no monetary compensa-tion, and the software used in the Real-Time Delphi wasprovided gratis.
Appendix A. Databases of terrorist and LoneWolf events
One of themost extensive database of terrorist events is theRAND Database of Worldwide Terrorism Incidents; it can beaccessed at http://smapp.rand.org/rwtid/search_form.php
This database permits filtering the recorded set by timeof occurrence, country, perpetrator, tactic, weapons, andseveral other parameters. Each incident is accorded aparagraph or so of description. There are 40,000 entriescovering events from 1968 to 2009; updating is reportedlyunderway. The database includes activities by both groupsand individuals but there is no filter to sort the LWs fromthe larger set.
The Global Terrorism Database (GTD) is even larger and ismaintained and made available by the National Consortium forthe Study of Terrorism and Responses to Terrorism (START) atthe University of Maryland, USA. It includes information onterrorist events around the world from 1970 through 2013 (tobe continuously updated), comprising more than 125,000cases. For each incident, it provides information on the dateand location of the incident, the weapons used and nature ofthe target, the number of casualties, and—whenidentifiable—the group or individual responsible. It is accessibleat http://www.start.umd.edu/gtd/
The New America Foundation's Database of homegrownextremism (at http://securitydata.newamerica.net/extremists/analysis) lists 122 incidents that include both LW and groupattacks—domestic and foreign—that have been performed byAmerican citizens and permanent residents that have occurredor have been foiled. It contains data from 2001 to 2014 (latestupdate in October, 2014) and focuses on and is organized bymethod of detection: e.g. tip, informant, suspicious activityreport. This is an excellent but limited set since it deals onlywith U.S. citizens and residents. While it is difficult to class theincident descriptions as being performed by a LWor not, a goodapproximation is that about 30% to 50% of the entries fit thedefinition of LW.
The Heritage Foundation compiles a list of foiled orfailed terrorist attacks that can be accessed at http://www.heritage.org/research/reports/2013/07/60-terrorist-plots-since-911-continued-lessons-in-domestic-counterterrorism.This listing includes 60 foiled events, mostly in the U.S.that have occurred between 9/11 and 2013. A review ofthis list shows that of the 60 attempts, some 22 (37%) wereby LWs.
The Wikipedia entry on Lone Wolf attacks lists 36 LWattacks conducted between 1998 and 2014. The definitionsused in this compilation are somewhat different; for examplethe list includes TimothyMcVeighwhodid not act totally alone.To the set of 36 examples provided one might add schoolshootings, the Sandy Hook shootings, Ivan Lopez another FortHood Shooter, the New York hatchet attack of October 13, 2014by Zale Thompson, the 2014 War Memorial killing in Canada,and several others.
Petter Nesser also compiled a list of 15 LW completedand failed terror attempts for further analysis, available at:http://www.terrorismanalysts.com/pt/index.php/pot/article/view/231/html. This list includes assassination attempts, plotsto blow up aircraft, arson, and mailing of toxic letters.
The report, Lone-Wolf Terrorism, A Case Study byThe European Research Consortium Transnational Terror-ism, Security and The Rule of Law (TTSRL); http://www.transnationalterrorism.eu/tekst/publications/Lone-Wolf%20Terrorism.pdf contains a number of important tables:
Chronology of lone-wolf terrorism in the United States,1940–2007 (that begins, interestingly enough with a reportof a series of bombings that tool place between 1940 and1956 in New York city)
Chronology of likely cases of lone-wolf terrorism in TTSRLsample countries, 1968–2007
Chronology of likely cases of lone-wolf terrorism in Canadaand Australia, 1968–2007
Incidents of terrorism in TTSRL sample countries, 1 January1968–1 May 2007 (lists 5646 incidents by country)
Hewitt, C. (2005) Political Violence and Terrorism in Modern
America (Westport and London: Praeger Security International)presents a chronology of terrorist events from 1956 to 2005,including “date of the incident, type of the incident, the groupor person responsible, where the attack occurred, and details ofthe act.”67
Harvey W. Kushner's (2003) Encyclopedia of Terrorism is anextraordinary 532 page compendium of terror incidents,locations, methods, and perpetrators with data to October2002.
67 Christopher Hewitt, Political Violence and Terrorism in Modern America: A
Chronology (Praeger Security International), September 30, 2005, ISBN-13:978-0313334184.
14 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
Gregory CopleyInternational Strategic Studies Association
Fig. 10. Sectoral demographics of the participants.
Fig. 9. Regional demographics of the participants.
15T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
Carol MasalaUniversity of the Federal Armed Forces MunichInstitute of Political ScienceNeubiberg, Germany
Ivana MilojevićMetafuture.orgSunshine Coast, Australia
Emilio MordiniCentre for Science, Society and CitizenshipRome, Italy
TomMurphyMedical DoctorThe Millennium ProjectDes Moines, Iowa, USA
Jean NegreanuFIRST GroupJerusalem, Israel
Pavel NovacekPalacyk UniversityOlomouc, Czech Republic
Concepcion OlavarrietaNodo Mexicano. El Proyecto Del MilenioMexico D.F. Mexico
Freddy PachysFIRST GroupAmit
16 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
Roman PeperhoveFreie Universität BerlinBerlin, Germany
Adrian PopNational University of Political and Administrative StudiesBucharest, Romania
Aaron RichmanTAM-CPhiladelphia, USA
Lawrence RobergeLaboure CollegeLudlow, MA, USA
Sheila RonisWalsh CollegeTroy, Michigan, USA
Shmuel C. ShapiraHadassah Medical OrganizationJerusalem, Israel
Yair SharanFIRST DirectorJerusalem, Israel
Gregory SimonsUppsala UniversityUppsala, Sweden
William TafoyaUniversity of New HavenWest Haven, CT, USA
Tim TepelUniversity of the Federal Armed Forces MunichInstitute of Political ScienceNeubiberg, Germany
Charles ThomasFutures Strategy GroupWest Hartford, CT, USA
Efrat TzadikMember of FIRSTKU Leuven UniversityLeuven, Belgium
Iuliana UdroiuNational University of Political and Administrative StudiesBucharest, Romania
Marius VacareluNational University of Political and Administrative StudiesBucharest, Romania
Madhur Varma
Columbia Asia HospitalsNew Delhi, India
Asher VaturiFirst GroupHod HaSharon, Israel
Jesús Alberto Velasco MataNodo Mexicano. El Proyecto Del MilenioMexico City, Mexico
Gregor WolbringUniversity of CalgaryCalgary, Canada
Yoel RabanTel Aviv UniversityTel Aviv, Israel
Simone Di ZioDepartment of Legal and Social SciencesUniversity “G. d'Annunzio”, Chieti-PescaraPescara, Italy
Ibon ZugastiPROSPEKTIKERSan Sebastián, Spain
Theodore Jay Gordon is a specialist in forecasting methodology, planning, andpolicy analysis. He is co-founder and Boardmember of TheMillennium Project,a global futures research think tank.Mr. Gordon started the consultingfirm, TheFutures Group in 1971. As CEO, his work involved technological innovation andforecasting, the design of analysis methodologies and the development ofstrategies—particularly strategies that promise to be productive in conditions ofhigh uncertainty. Prior to forming The Futures Group, Mr. Gordon was one ofthe founders of The Institute for the Future. He performed early research onthe Delphi method at RAND Corporation where he co-authored the first largescale Delphi study. Throughout his career he has been at the forefront ofdevelopment of forecasting and analysis methodology including the CrossImpactMethod, Trend Impact Analysis, and the State of the Future Index (SOFI).He was Regents Professor at the UCLA Graduate School of Business (1968), anda Woodrow Wilson scholar at Bowdoin (1975). He has lectured at manyuniversities in business, planning, innovation, forecasting and engineering. Heis the recipient of the Ed Cornish “Futurist of the Year” award and is a recipientof the Shaping Tomorrow Lifetime Achievement Award.
Dr. Yair Sharan is currently the director of TAM-C/FIRST group active in thesecurity and technology field and a senior associate researcher in Begin-SaadatCenter for strategic studies (BESA) in Bar Ilan University in Israel. He is a Col(res) in the IDF. He has been the director of the Interdisciplinary Center forTechnological Analysis and Forecasting (ICTAF) at Tel Aviv University Israel. Heearned his Ph.D. in Physics from the Weizmann institute of Science. He is theCo-Director of the Israeli node in the Millennium Project. His main fields ofinterest are research and science policy, the impact of technology on nationalstrategy, technology foresight and security foresight, technology assessmentand more. He coordinated the EU project FESTOS on emerging security threatsand PRACTIS on the future of privacy. Published studies on topics like “NonConventional Terrorism”, “Psychological Deterrence”, “Emerging Technologies”and “Future Threats of Terrorism”. Current studies include “Lonely wolf — anemerging terror threat”, “Synthetic Biology — Risks and Chances”, “emergingtechnologies and their impact on security and privacy”.
Elizabeth Florescu is a Director of Research at TheMillennium Project, a globalfutures research think tank. She has beenworkingwith TheMillenniumProjectsince 1997. Elizabeth has a vast experience in futures research, economicmodeling, and diplomacy and security-related domains. She is co-author of theannual “State of the Future” reports, the “Environmental Crimes, MilitaryActions, and the International Criminal Court”, and several articles, and invited
17T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
reviewer at specialized magazines. She has been one of the principalinvestigators working on the monthly environmental scanning reports for theU.S. Army Environmental Policy Institute, assessing worldwide environmentand new technologies—related issues that might trigger future changes tointernational regulations with potential implications for the military. She
has been a speaker at conferences in North America, Europe, and Africa, and isinvolved in committees and forums addressing issues related to S&T,environment, security, international regulations, ontology, and knowledgemanagement.
18 T. Gordon et al. / Technological Forecasting & Social Change xxx (2015) xxx–xxx
Please cite this article as: Gordon, T., et al., Prospects for Lone Wolf and SIMAD terrorism, Technol. Forecast. Soc. Change (2015),http://dx.doi.org/10.1016/j.techfore.2015.01.013
