Mr. Chairman, Senator Lugar, and Members of the Committee: I thank you for the opportunity to appear before you today to discuss the threat of al Qaeda in Afghanistan and beyond.

Al Qaeda is once again at the forefront of U.S. Government policy debate. Our strategic interest in Afghanistan is linked to the protection of the homeland and that of our Western allies against terrorist attacks. A moment?s reflection will demonstrate this. Al Qaeda found sanctuary in the Sudan for four years, from 1992 to 1996, when the Sudanese government expelled it. During this Sudanese phase, al Qaeda developed its strategy to target the West, and especially the United States and trained potential terrorists there. Indeed, the planning of the simultaneous bombings of our Embassies in Nairobi and Dar es Salaam was done in Khartoum. Had al Qaeda not been thrown out of the Sudan, I have no doubt that we would be discussing strategy options about the Sudan rather than Afghanistan.

Our ultimate goal of homeland security will be served through a better understanding of the threat confronting it in order to “disrupt, dismantle, and eventually defeat al Qaeda and its allies.” Let me describe this global threat through a comprehensive survey that I conducted of all the al Qaeda plots in the West, all the al Qaeda affiliate plots in the West and all the plots done “in the name of al Qaeda” in the West since the formation of al Qaeda in August 1988. It is necessary to expand our inquiry because al Qaeda is now only one of the many actors in this global neo-jihadi terrorist threat against the West. I call it neo-jihadi because the terrorists have appropriated this contested concept to themselves much to the protest of respected Islamic scholars and the mainstream Muslim communities worldwide[1]. Terrorism for the purpose of this project is the use of violence by non-state collective actors against non-combatants in the West in pursuit of a self-appointed global jihad.

I conducted this survey when I spent a year at the U.S. Secret Service and an additional year at the New York Police Department as its first scholar-in-residence. Although both organizations helped me immensely, the following remarks are my own and cannot be read as their position or opinions. Because homeland security in the West essentially means population protection in the West, I have limited the inquiry to violent plots to be executed in the geographical territory of the West. By the West, I mean North America, Australia and Western Europe, with the exception of the civil war in the Balkans since terrorism is often a tactic of war, but wartime terrorism may not teach us much about terrorism during peace time. To be included in the survey, each plot had to have some loose operational or inspirational link to al Qaeda or its affiliates; it had to reach a certain level of maturity, characterized by overt acts in furtherance; it consisted of violent acts targeting people in the West, and therefore excluded cases of purely financial or material support for terrorist acts committed elsewhere; some planning had to be done in the West; and terrorists had to initiate the plot. To accurately evaluate the threat, I of course included both successful and unsuccessful plots, which are the true measure of the extent of the threat, rather than just the successful ones. The global neo-jihadi terrorist threat includes plots under the control of al Qaeda core; al Qaeda affiliates like the Algerian Groupes Islamiques Armes (GIA), Pakistani Lashkar e-Toyba (LT), the Uzbek Islamic Jihad Union (IJU), the Pakistani Tehrik e-Taliban Pakistan (TTP)…; and threats by autonomous groups inspired by al Qaeda like the Dutch Hofstad group. I excluded lone wolves, who were not physically or virtually connected to anyone in the global neo-jihad, for they often carry out their atrocities on the basis of delusion and mental disorder rather than for political reasons.

My sources of information were legal documents, trial transcripts, consultations with foreign and domestic intelligence and law enforcement agencies, to which my position gave me access. Although all these plots are within the open source domain, I did corroborate the validity of the data in the classified domain.

The specified criteria yielded a total of 60 global neo-jihadi terrorist plots in the West, perpetrated by 46 terrorist networks in the past two decades, from the first World Trade Center attack on February 26, 1993 to the December 16, 2008 arrest of Rany Arnaud, who was plotting to blow up the Direction Generale du Renseignement Interieur, the French FBI equivalent, in a suburb of Paris. Although people associate al Qaeda plots with airplanes or bombs, the plots were quite diverse: simple assassinations, attempted kidnapping and decapitation, car/truck bombs, airplane hijacking, and improvised explosive devices. Some operations were suicidal, but most were not. Of all the plots, only one is completely unsolved – the bombing of the Port Royal Metro station in Paris on December 3, 1996, which resulted in many casualties. Although completely unsolved, the timing, context and mode of operation seem to point to the GIA, trying to avenge its followers, who were put on trial around that time.
The following graph is the timeline distribution of the plots.

We can see from the above graph that global neo-jihadi terrorist plots preceded the 9/11/01 attack when the Western public first started to appreciate the true extent of the threat confronting it. The first plot in the West was the first World Trade Center bombing in February 1993, or about four and a half years after the creation of al Qaeda proper. The timeline distribution of the plots is bi-modal. The first peak consisted of raids by the Algerian GIA against France and stopped in 1996; the later plots were more widely geographically distributed and reached a peak in 2004, after which they declined. In the recent controversy over whether al Qaeda (however defined, here I am using a more inclusive and therefore much wider definition of the threat in the West) is on the move or on the run, we can see that the wider “al Qaeda” threat or the global neo-jihadi terrorist threat is definitely on the run since its high water mark of 2004.

Some networks of terrorists, who temporarily escaped arrest, carried out multiple plots in the West. This is especially true of the 1995 wave of ten GIA plots against France, carried out by the same network in France. In order to understand the actual threat, as opposed to the inability of local police forces to disrupt existing networks, I also coded the global neo-jihadi threat to the West according to the specific terrorist networks carrying out operations (as opposed to plots). Coding the data according to networks rather than plots gives the following graph.

What is most affected by this coding is the collapse of the GIA wave of bombings in France in 1995, now represented by the same group rather than the ten separate plots. Again, loosely global neo-jihadi networks in the West preceded the 9/11/01 operation. Here, the graph indicates that global neo-jihadi networks in the West became more numerous in 2001, experienced a temporary small decline, and reached its 2004 high water mark, after which it declined, especially after 2007. So, here again, “al Qaeda” is on the run and not on the move. I suspect the post 2003 bump in the number of networks threatening the West in the name of AQ was a reaction to the Western invasion of Iraq.

Although the press likes to call any militant Islamist plot an al Qaeda plot, let us see how many are truly al Qaeda plots. I coded the command and control of each plot according to the following classification (I did not code the 1996 Paris Metro plot because it is still unsolved):

• AQ Core means that AQ proper directed and controlled the operation.
• AQ Affiliated means that an international terrorist organization affiliated with AQ, such as LT or IJU, directed and controlled the operation.
• AQ Inspired means that there was no direction or control by any of the above organization for the plot. In other words, the plot was completely autonomous.

In this coding system, I leaned backward to give credit to a terrorist organization when there was any doubt about its command and control over an operation. I did this to increase the probability of detecting any coordination of global neo-jihadi terrorism by a single entity, a sort of neo-jihadi equivalent of the Comintern – the Soviet Central Committee in Moscow that tried to coordinate Communist activities worldwide.

The result is:

• 12 AQ Core controlled operations (20%)
• LAX millenial plot (1999)
• Strasbourg Christmas Market bombing plot (2000)
• 9/11/01 attack (2001)
• Paris Embassy bombing plot (2001)
• Belgian Kleine Brogel US Air Force base bombing plot (2001)
• Shoe bomber plot (2001)
• London fertilizer bomb plot (Operation Crevice, 2004)
• London limousine bombing plot (Operation Rhyme, 2004)
• London 7/7 bombings (Theseus case) (2005)
• London 7/21 bombing plot (Vivace case) (2005)
• London airplanes liquid bomb plot (Operation Overt) (2006)
• Danish Glasvej bombing plot (Operation Dagger) (2007)
• 15 AQ affiliated terrorist organizations controlled operations (25%)
• 11 GIA plots against France (1994-5)
• German al-Tawhid bombing plots (Zarqawi group) (2002)
• Sydney bombing plot (Brigitte-Lodhi, LT controlled) (2003)
• German Sauerland bombing plot (IJU controlled) (2007)
• Barcelona bombing plot (alleged TTP control) (2008)
• 32 AQ inspired terrorist plots, carried out either on behalf of al Qaeda or other transnational terrorist organizations (54%)

Al Qaeda-inspired autonomous plots constitute the majority of all the plots, followed by al Qaeda affiliated plots, with true al Qaeda plots closing out the sample at only 20%. Viewing the graph chronologically, al Qaeda Core did not start this terrorist campaign against the West. Indeed, all al Qaeda Core plots in the West took place after bin Laden?s 1998 hukm (his „considered judgment,? not fatwa as is incorrectly reported in the West and which carries much less authority than a fatwa)[2]. Two attacks in New York City conducted by former Afghan Arabs inaugurated this worldwide wave of bombings against the West. They were conducted locally, and there is no evidence that there was any guidance, direction or control by al Qaeda Core. If anything, they were more closely connected with the Egyptian Islamic Group than al Qaeda or its ally, the Egyptian Islamic Jihad. These attacks were followed by a large wave of GIA attacks against France, which had to do with the internal dynamics of the Algerian civil war in the 1990s, and again had no guidance, direction or control from al Qaeda Core. The actual al Qaeda Core plots in the West began in late 1999, as part of a wave of worldwide bombings to mark the dawn of the new Western Millennium, peaked in 2001, and decreased thereafter to about one plot per year, with a small uptick in 2004-2005 and fading over the next two years. Despite even recent claims that al Qaeda is on the move, it is clear that al Qaeda in the West has been on the decline since its apogee of 2001. When studying a phenomenon, it is important to count and look at the trend. When one relies on out of context anecdotal evidence, it is easy to make mistakes. I suspect that the recent advocates for a “resurgent” al Qaeda were confused by the complexity of the 2006 London airplanes liquid bomb plot (Overt case) and mistook complexity for resurgence. The fact is clear that since its loss of sanctuary in Afghanistan in 2001, al Qaeda proper has had trouble projecting to the West. It was able to operate locally in South Asia and Iraq, especially after al Zarqawi proclaimed a merger of his organization with al Qaeda.

Let’s look at the past five years:

• 6 AQ Core plots (2004 Rhyme and Crevice plots; 2005 Theseus and Vivace cases; 2006 Overt case, all in Britain, and 2007 Dagger plot in Denmark)

• 2 AQ Affiliated plots (2007 Sauerland & 2008 Barcelona Plots)

• 25 AQ Inspired autonomous plots, conducted by homegrown perpetrators, with no connections whatsoever with any formal transnational terrorist organizations

The above statistics are crystal clear: 78% of all global neo-jihadi terrorist plots in the West in the past five years came from autonomous homegrown groups without any connection, direction or control from al Qaeda Core or its allies. The „resurgent al Qaeda? in the West argument has no empirical foundation. The paucity of actual al Qaeda and other transnational terrorist organization plots compared to the number of autonomous plots refutes the claims by some heads of the Intelligence Community (Hayden, 2008) that all Islamist plots in the West can be traced back to the Afghan Pakistani border. Far from being the “epicenter of terrorism,” this Pakistani region is more like the finishing school of global neo-jihadi terrorism, where a few amateur wannabes are transformed into dangerous terrorists.

The graph also shows a sporadic involvement of al Qaeda affiliated terrorist groups in plotting against the West in the past six years. These groups located in Pakistan are showing an increased ability to project against the West, although most of their operations are still confined to South Asia. However, in the internal rivalry among terrorist groups in South Asia, the quickest way to establish one?s reputation is to demonstrate an ability to strike in the West. Although it is rare for al Qaeda core to claim credit for its operations in the West, its rivals in South Asia have been quick to claim credit, even for failed plots. The Islamic Jihad Union claimed credit for the failed Sauerland group plot in September 2007 and Baitullah Mehsud, the deceased chief of Tehrik-e Taliban Pakistan claimed credit for the failed Barcelona Plot of January 2008 – although this last claim must be taken with a great deal of caution because he has claimed credits for mishaps in the West that had nothing to do with his organization, like the power outage in the U.S. Midwest in 2007 and the mass murder incident in Binghamton, New York on April 3, 2009. These empty self-promotions have been categorically refuted by federal authorities. The West may well find itself caught in this militant rivalry for global neo-jihadi supremacy.

My coding probably overestimated the importance of formal terrorist groups. Most of the recent plots coded as under al Qaeda command and control, like the 2004 London fertilizer bomb plot, did not involve such frequent communication with al Qaeda, but included instead a short meeting with a high level representative of al Qaeda, where local Western terrorist wannabes informed al Qeada representatives, Abdal Hadi al Iraqi and his lieutenant, of their own initiative to conduct operations in the West. In such cases, it seems that the meeting with al Qaeda leadership did not affect the desire of the local terrorists to conduct such operations. Here the role of the al Qaeda was passive agreement with little influence on the plot.

The dramatic increase in global neo-jihadi terrorism in the first decade of the 21st Century has come from al Qaeda inspired autonomous groups with no link to formal transnational terrorist groups. This is especially true since the 2003 invasion of Iraq, which has inspired local young Muslims to strike out against the West. It seems clear that this invasion has created more terrorists in the West, refuting the thesis that “we are fighting them there, so we don?t have to fight them here.” The fact that these plots peaked in 2004, one year after the invasion of Iraq provides empirical support linking the two events. These scattered plots, not coordinated by any central terrorist body and constituting almost 80% of the plots against the West in the past five years, illustrate how the threat against the West is degenerating into a “leaderless jihad.”[3] Far from being directed by a Comintern, global neo-jihadi terrorism is evolving to the structure of anarchist terrorism that prevailed over a century ago, when no such global coordinating committee was ever found despite contemporaneous belief in its existence.

Within this cluster of al Qaeda inspired autonomous groups is a troubling emerging pattern of lone wolves, directly linked via the Internet to foreign al Qaeda affiliated terrorist organizations: the 2004 Rotterdam Plot (Yehya Kadouri), the 2007 Nancy plot (Kamel Bouchentouf), the 2008 Exeter plot (Nicky Reilly) and the 2008 French Direction Centrale du Renseignement Interieur plot (Rany Arnaud). Although these young men are willing to sacrifice themselves for these affiliate terrorist groups, they have never met them face to face. This may become a trend that will increase in the future.

Another dimension of allied al Qaeda involvement in plots against the West is financial support of these plots. Again, in examining each global neo-jihadi terrorism network for such support, I have erred on the side of inclusiveness of al Qaeda support in this coding scheme.

Out of forty-five global neo-jihadi terrorist networks in the West, al Qaeda at least partially funded ten. But this overstates its importance in this regard. The funding of the 1993 World Trade Center plot was minimal, and consisted of Khalid Sheikh Mohammed sending a few hundred dollars to his nephew Ramzi Yousef. It is unclear where the money came from, but for the sake of this study, let us assume it came from al Qaeda. The same goes for the GIA wave of bombings in France in 1995. Bin Laden funded the Al Ansar newsletter in London via Rachid Ramda, who funded the bombing campaign. I do not know where the money for this campaign (as opposed to the newsletter) came from. I suspect that it came from the GIA itself through its fund raising campaign throughout Europe. However, let us again assume that it came from bin Laden either directly or indirectly.

We can see that from 1999 to 2001, al Qaeda either partially or fully funded its operations against the West. This was either in the form of seed money ($10,000 given to Ahmed Ressam for the 1999 LAX bombing plot or the 2000 Strasbourg Christmas Market bombing plot). In each case, the perpetrators were supposed to supplement their initial funds via their own means (robbery in Ressam?s case; drug sales for the other). Sometimes, the funding was paid in full, as in the 9/11/01 plot. I assume that al Qaeda at least partially funded the rest of the 2001 al Qaeda plots since I came across no evidence that these perpetrators raised any money on their own. The two alleged al Qaeda plots in 2005 were a departure from this pattern, as there is no evidence that the two London bombing plots of July 2005 received any money from al Qaeda. The last alleged al Qaeda plot, the Danish Glasvej (Dagger) case indicates that the main perpetrator, Hamad Khurshid, came back from Pakistan with $5,000 in cash. It is true that, except for the 9/11/01 operation, terrorist plots are not expensive to carry out. Autonomous terrorists had no choice but to raise the funds for their operation themselves.

On the other hand, the al Qaeda-affiliated transnational terrorist groups seemed to have funded their own operations. The GIA plots were fully funded from outside and none of the perpetrators were tasked with raising money for the plots. The 2002 German al Tawhid plot was probably funded by Zarqawi. LT funded the Sydney plot through money transfers to Willie Brigitte in 2003, and the IJU seemed to have funded the 2007 German Sauerland plot. It is unknown the degree of financial support that the potential perpetrators of the 2008 Barcelona plot received from Mehsud?s organization.

For those who like to follow the money, only a very few plots have been funded from the outside in the past five years. Of the twenty-nine global neo-jihadi terrorist networks involved during that period, Al Qaeda core funding has been implicated in only two – Hamad Khurshid and the London Rhyme case. Even if we add the non-al Qaeda funded Sauerland case and possibly the TTP Barcelona case, the total increases to only three or four out of twenty-nine cases (10% or 14%). Since the money involved was mostly in the form of cash, following the trail of money will not detect global neo-jihadi terrorism plots in the West. The vast majority these networks in the past five years have raised their own money.

It has been argued that training by a formal terrorist organization is critically important because it transforms amateurs into seasoned terrorists. Several Western intelligence leaders have stated that all significant global neo-jihadi terrorist plots lead back to the Federally Administered Tribal Areas of Pakistan (FATA). The next graph tests this claim. I plotted the overseas training for all the terrorist networks and coded them as receiving training from al Qaeda, an al Qaeda affiliate, or no training at all – just al Qaeda inspired. Again, I erred on the side of over-inclusiveness of such training, even if just one person in the network, who might not have been involved in the planning of the plot, had simply undergone familiarization training, which did not teach any significant bomb making skills. For this graph, I coded Bouyeri as being separate from the Hofstad network because he carried out the assassination of Theo van Gogh on his own in 2004 and had not gone to any training camp

Out of 46 different networks attempting terrorist operations in the West,

• 16 had at least one member that underwent training at an AQ Core facility (35%)
• 10 had at least one member that underwent training at an AQ affiliated facility (22%)
• 20 had no training at all (43%)

Lumping the data together hides some important trends. First, more people have trained from al Qaeda and al Qaeda affiliates than are under the control of these respective organizations. Lately, in the press and perhaps the intelligence community, there is a presumption that attendance in a formal terrorist organization training camp is equivalent to being under control of that organization. So, I included the 2004 London fertilizer plot (Operation Crevice) and the two 2005 London underground bombing plots as al Qaeda controlled because the perpetrators had allegedly received al Qaeda training. However, there was no evidence of extensive communication between the perpetrators in the field and al Qaeda Core in Pakistan, unlike the 9/11/01 plot or the 2006 London airplanes liquid bomb plot, where the perpetrators were in almost daily communication with al Qaeda core, or the 2007 Sauerland plotters, who were in constant e-mail contact with their IJU sponsors.

This equation of training camp attendance with foreign terrorist organization control was not presumed for the pre-2001 plots, when attendance in an al Qaeda camp did not mean al Qaeda control. For example, Ramzi Yousef, the bomb maker for the first World Trade Center bombing in 1993, never belonged to al Qaeda, but had undergone extensive training at al Qaeda funded camps and had taught at Abdal Rabb Rasul Sayyaf?s University of Jihad. Likewise, members of the 2002 al Tawhid plot had been trained at al Qaeda camps before joining Abu Musab al-Zarqawi?s al Tawhid organization. Again, the two ricin plots (the 2002 French Chechen network and the 2003 British ricin plot [Operation Earth]) included members who had trained in al Qaeda camps, even though neither plot seemed to have been known or sanctioned by al Qaeda as far as I know.

Al Qaeda funded most of the training camps in Afghanistan before the U.S. invasion in the fall of 2001. Anyone who had traveled to Afghanistan for training at that time was bound to have been trained in an al Qaeda funded camp. The cases just cited included members who had been in Afghanistan before the fall of the Taliban regime. The result was that graduates from al Qaeda camps in the 1990s dominate global neo-jihadi terrorism from 1999 to 2002. By the time they were planning their operations in the West in 2002 or 2003, they no longer had any active link to al Qaeda. Since 2002, al Qaeda trained terrorists averaged just one plot a year.

As the availability of al Qaeda training faded over time, al Qaeda affiliated terrorist organizations in Kashmir or the Federally Administered Tribal Areas of Pakistan, such as Laskhar-e Toyba or the Islamic Jihad Union, began to fill in the gap starting in 2003 and the graduates of their camps also average about one plot a year. So, while terrorist networks that had training dominate the overall sample (57%), this trend has been reversed in the past five years as only 40% had such training. Indeed, all those who underwent training in the past five years, acquired it in Pakistan, not Afghanistan.

Although I use the generic term “training camp” to describe the place of training before and after 2001, the meaning of the term has since changed dramatically and overestimates the formality and sophistication of training received by global neo-jihadi terrorist networks in the West after 2001. Gone are the large formal camps like Khalden, Farooq or Darunta in Afghanistan, which could accommodate hundreds of novices and had a formal curriculum with increased levels of sophistication sometimes lasting up to a year for the select few (see Ahmed Ressam?s training for the 1999 LAX millennium plot). After the 2001 U.S. invasion of Afghanistan, these formal training facilities were destroyed. People traveling to Pakistan afterwards either went to formal training facilities conducted by Kashmiri terrorist groups in Kashmir (see the legal judgment on Willie Brigitte for a description of such camps) or had to arrange for their training through hiring of a private trainer. These new “camps” were nothing like the former ones: they were small rented housing compounds or even two tents in a goat patch, where one instructor and his son gave private lessons to at most a dozen students, who directly paid for their instruction, the duration of which could be as short as two days to about three weeks (see the transcripts of the 2004 Crevice or the 2005 Theseus cases, which describe this process).

Later, after a series of truces signed between FATA tribal leaders and the government of Pakistan between 2004 and 2006, al Qaeda or IJU provided more formal training in Waziristan, but they never reached the level of sophistication in instruction that prevailed before 2001. These new facilities in Waziristan were more visible than before and could accommodate up to about twenty trainees at a time. Indeed, the presence of these camps probably led to alarms that al Qaeda was resurgent. Strangely enough, the presence of these new “camps” did not affect the frequency of al Qaeda linked plots in the West. The slight bump in frequency of terrorist trained arrests or actual bombings in 2004 and 2005 was not due to these truces, because the training of the perpetrators preceded the truce agreements. Despite the widespread alarms in the West, the truces do not appear to have any effect on global neo-jihadi terrorism in the West.

In any case, the graph shows clearly that the majority of global neo-jihadi terrorist networks from 2004 onwards did not have any formal training from foreign terrorist groups (60%), contrary to the statements of Intelligence agency chiefs on both sides of the Atlantic. They were purely homegrown and had no link to the FATA, which some have called “the epicenter of terrorism.” Instead, they had to rely on themselves and the Internet for their acquisition of terrorist skills, consistent with the leaderless jihad argument.

How dangerous is global neo-jihadi terrorism? In other words, what is the result of global neo-jihadi terrorist plots in the West? I coded all 60 plots in the West in terms of whether they caused any injuries; were carried out but failed (no explosion because of a technical error); or were interrupted through law enforcement arrests.

The results are as follows:

• 14 Plots were successful in terms of incurring any injury and or death (23%)

• Only 2 al Qaeda core plots in the West in the past two decades were successful (9/11/01 and 7/7/05). Of course, they were among the most devastating, resulting in about 3,000 fatalities for 9/11 and 52 fatalities for 7/7.

• 9 were GIA plots against France, from 1994 to 1996 (I have counted the 1996 Paris Port Royal metro station bombing in this total. The total for all of these attacks is 17 fatalities)

• 3 were al Qaeda inspired plots (1993 World Trade Center bombing, resulting in 6 fatalities; 2004 Madrid bombing, resulting in 191 fatalities; 2004 Bouyeri?s assassination of Theo van Gogh)

• 10 Plots resulted in failure to explode (17%)

• 3 failures in networks that had succeeded elsewhere (2 by 1995 GIA network in France; and by 2004 Madrid network when bomb on the AVE train line near Toledo failed to detonate)

• 2 failures by al Qaeda trained networks (2001 Shoe bomb plot and 7/21/05 London underground bombing plot)

• 1 failure in network of French Bosnian war veteran (Roubaix group)

• 4 failures in networks that had no foreign terrorist organization training (2004 Rotterdam plot; 2006 Koblenz train plot; 2007 Doctors‟ plot; and 2008 Exeter bomb plot)

• 36 Plots were interrupted through arrests (60%)

It is interesting to note that for all the fear of al Qaeda, the organization managed only two successful plots in the West in the last twenty years! The fact that they were so deadly overshadows this truth. Indeed, successful independent plots outnumber successful al Qaeda plots in the West. However, both are eclipsed by the GIA, which infiltrated a team of trained terrorists to France, whose wave of terror in the mid-1990s accounts for almost two thirds of all successful global neo-jihadi bombings.

This low rate of success (23%) should not be much comfort to intelligence or law enforcement agencies. In ten plots, the terrorists succeeded in setting their bombs down without being detected. The bombs simply did not detonate, which cannot be due to good intelligence or police work. So, the rate of a plot going to termination without being detected is 40%, a very high rate indeed, no cause for comfort. Lest the reader thinks that the cause for failure to detonate was the lack of training by homegrown wannabes, six out of the ten failures happened to groups that had been trained or been successful before. So, 60% of the failures to detonate were not due to poor training but to poor execution by experienced terrorists.

It appears that either we are getting luckier or this terrorist threat is diminishing. In the United States, the last casualty dates back eight years to 9/11/01. There has not been even one plot that went to termination since then. In the rest of the West, there has not been a single casualty in the past four years. The last casualty dates back to 7/7/05, the first London underground plot. However, in the past four years, Europe has witnessed a series of bombs that failed to detonate: the 2005 second London underground plot (Vivace case); the 2006 German Koblenz trolley bombs; the 2007 London and Glasgow Doctors‟ plot; and the 2008 Exeter bomb plot by Nicky Reilly. The last three plots have no physical link to any transnational terrorist groups.

How effective is formal terrorist training for the successful completion of a plot? Several critics have tried to downplay the recent surge of autonomous homegrown plots as less dangerous than those of formally trained terrorists. I analyzed the results of global neo-jihadi terrorist networks according to their type of training: al Qaeda core training; al Qaeda affiliates‟ training; or no formal training at all (al Qaeda inspired). Excluding the unsolved 1996 Paris Port Royal metro bombing because of lack of information, this leaves forty-five networks. But an untrained member of the Hofstad network, Mohammed Bouyeri, carried out a successful assassination on his own. His “trained” colleagues, Jason Walters and Ismail Akhnikh, had not been aware of his plan and provided no guidance or help. Therefore, I decided to code Bouyeri‟s assassination of Theo van Gogh as a separate network, and as al Qaeda inspired. The results are the following:

• 16 AQ Core trained networks:
• 3 succeeded (1993 World Trade Center bombing; 9/11/01; and 7/7/05 London underground bombing) [19%]
• 2 failed to explode (2001 Shoe bomber; 7/21/05 London underground plot)
• 11 were detected and arrested before hand
• 10 AQ Affiliate trained networks
  • 2 GIA networks succeeded (1994 AF hijack; 1995 wave of bombing in France) [20%]
  • 1 failed to explode (1996 Lille plot)
  • 7 were detected and arrested beforehand (including Hofstad network)
• 20 AQ Inspired networks (no formal training)
• 2 succeeded (2004 Madrid bombings & 2004 Bouyeri assassination of van Gogh)[10%, but only 5% if we don‟t count the assassination, which requires no training]
• 3 failed to explode
• 16 were detected and arrested beforehand.

The above results seem to indicate that formal training matters. Both al Qaeda core and al Qaeda affiliate formal training resulted in an approximate success rate of 20%, while lack of training led to a success rate of 10%. So, training doubles the probability of success in a terrorist network. However, if the assassination of Theo van Gogh is eliminated from the sample, the resulting rate of success of the untrained networks falls to 5%. In this case, training would quadruple the probability of success in a terrorist network.

Viewing the sample as a whole obscures the degradation of the importance of training in the past five years. During this period, of twelve trained terrorist networks, only one succeeded in causing any casualty, the 7/7/05 London underground bombing. Two untrained networks out of sixteen succeeded in inflicting casualties: the 2004 Madrid bombing – where the bombers got access to dynamite, det-cord and detonators, and did not have to manufacture their explosive – and the 2004 Bouyeri assassination of van Gogh.

I am sorry to have been so lengthy in the presentation of the survey, but the devil is in the empirical details to escape another round of hysterical rhetoric so common in discussion of global neo-jihadi terrorism. Now that I‟ve laid down the facts, let me address some of the unexamined assumptions, myths and misconceptions about the “al Qaeda threat” in Afghanistan and beyond.

1. The threat to the West has unfortunately expanded beyond al Qaeda per se. The various terrorists attempting to carry out operations in the West for al Qaeda allies or in its name clearly outnumber al Qaeda operations. In the past five years, al Qaeda core has been responsible for only 18% of these plots. 78% of these plots during this period have been carried out by homegrown terrorists, inspired by al Qaeda, but with no connection with any formal transnational terrorist organization – evolving into a Leaderless Jihad. This survey does not include the new al Shabaab threat to the West, which has too recently surfaced to be included. But it stems from Somalia and not Afghanistan.
2. The dichotomy of the present policy options between counter-terrorism and counter-insurgency is a false one. The choice is not between counter-terrorism and counter-insurgency, but between counter-terrorism and counter-terrorism plus counter-insurgency. No matter what happens in Afghanistan, all Western powers will continue to protect their homelands with a vigorous counter-terrorism campaign against al Qaeda, its allies and its homegrown progeny. The policy option really boils down to, what is the added value of counter-insurgency in Afghanistan to a necessary and continuing counter-terrorism strategy worldwide?
3. The proposed counter-insurgency strategy in Afghanistan is at present irrelevant to the goal of disrupting, dismantling and defeating al Qaeda, which is located in Pakistan. None of the plots in the West has any connection to any Afghan insurgent group, labeled under the umbrella name “Afghan Taliban,” be it a part of Mullah Omar‟s Quetta Shura Taliban, Jalaluddin Haqqani‟s Haqqani Network, or Gulbuddin Hekmatyar‟s Hezb-e Islami. There has not been any Afghan in al Qaeda in the past twenty years because of mutual resentment between al Qaeda foreigners and Afghan locals. In the policy debate, there is an insidious confusion between Afghan Taliban and transnational terrorist organizations. Afghan fighters are parochial, have local goals and fight locally. They do not travel abroad and rarely within their own country. They are happy to kill Westerners in Afghanistan, but they are not a threat to Western homelands. Foreign presence is what has traditionally unified the usually fractious Afghan rivals against a common enemy. Their strategic interest is local, preserving their autonomy from what they perceive as a predatory corrupt unjust central government. They do not project to the West and do not share the internationalist agenda of al Qaeda or its allied transnational terrorist organizations.
4. The second prong of the proposed counter-insurgency strategy in Afghanistan is the prevention of al Qaeda‟s return to Afghanistan through a military surge. The assumption is that the return to power by the Taliban will automatically allow al Qaeda to reconstitute in Afghanistan, complete with training camps and resurgence of al Qaeda‟s ability to project to the West and threaten the homeland.
1. The possibility of Afghan insurgents winning is not a sure thing. Twenty years ago, it took a far better armed and far more popular insurgency more than three years to take power after the complete withdrawal of Soviet forces from Afghanistan. Unlike 1996, when the Taliban captured Kabul, the label Taliban now includes a collection of local insurgencies with some attempts at coordination on a larger scale. The Taliban is deeply divided and there is no evidence that it is in the process of consolidating its forces for a push on Kabul. Local Taliban forces can prevent foreign forces from protecting the local population, through their time honored tactics of ambushes and raids. General McChrystal is right: the situation in the countryside is grim. But this local resistance does not translate into deeply divided Taliban forces being able to coalesce in the near future into an offensive force capable of marching on to Kabul. Command and control frictions and divergent goals hamper their planning and coordination of operations. They lack popular support and have not demonstrated ability to project beyond their immediate locality.
2. Taliban return to power will not mean an automatic new sanctuary for al Qaeda. First, there is no reason for al Qaeda to return to Afghanistan. It seems safer in Pakistan at the moment. Indeed, al Qaeda has so far not returned to Taliban controlled areas in Afghanistan. Al Qaeda‟s relationship with Taliban factions has never been very smooth, despite the past public display of Usama bin Laden‟s pledge of bayat to Mullah Omar. Al Qaeda leaders seem intimately involved in the Haqqani network in North Waziristan, less so with Mullah Omar‟s Quetta Shura, and even less with Gulbuddin Hekmatyar‟s forces. Indeed, the presence of al Qaeda in Afghanistan divided Taliban leaders before their downfall. Likewise, loyalty for Taliban leader Mullah Omar also divided al Qaeda leadership. This complex relationship between al Qaeda and Afghan Taliban factions opens up an opportunity for the U.S. Government to mobilize its political savvy based on a deep understanding of local history, culture and politics to prevent the return of a significant al Qaeda presence in Afghanistan[4].
3. Even if a triumphant Taliban invites al Qaeda to return to Afghanistan, its presence there will look very similar to its presence in the FATA. Times have changed. The presence of large sanctuaries in Afghanistan was predicated on Western not so benign neglect of the al Qaeda funded camps there. This era is gone because Western powers will no longer tolerate them. There are many ways to prevent the return of al Qaeda to Afghanistan besides a national counter-insurgency strategy. Vigilance through electronic monitoring, spatial surveillance, a networks of informants in contested territory, combined with the nearby stationing of a small force dedicated to physically eradicate any visible al Qaeda presence in Afghanistan will prevent the return of al Qaeda in Afghanistan. The proper military mission in Afghanistan and elsewhere is sanctuary denial.
5. Counter-terrorism is working. The escalation from a more limited and focused counter-terrorism strategy to a larger combined counter-terrorism and counter-insurgency strategy (in a country devoid of the al Qaeda presence!) is predicated on the assumption that the terrorist threat is either stable or increasing – meaning that counter-terrorism has failed. The timeline graphs clearly show that the threat is fading, from its high water mark of 2004. There has been no global neo-jihadi terrorist casualty in the United States in the past eight years and none in the West in general in the past four years. Of course, al Qaeda is not dead as long as its top leadership is still alive. This cannot be attributed to a loss of intent from al Qaeda and its militant rivals. From all indications, including recent debriefs of terrorist wannabes captured in Pakistan and the West, the respective leaders of global neo-jihadi terrorism are still enthusiastically plotting to hit the West and do not hesitate to proclaim their desire on the Internet. Nor is this due to the counter-insurgency in Afghanistan because al Qaeda and its allies all have their training facilities in Pakistan. It is due to effective counter-terrorism strategy, which is on the brink of completely eliminating al Qaeda. A dead organization will not be able to return to Afghanistan.
6. The reasons for the effectiveness of the counter-terrorism strategy so far are multiple. First and foremost is al Qaeda‟s inability to grow. Unlike the pre-9/11/01 period, al Qaeda leaders have generally not incorporated new recruits among its ranks. The leadership of al Qaeda still harks back to the fight against the Soviets in the 1980s. Because he has been hiding full time, Osama bin Laden has not been able to appoint and train a new group of top leaders and there is no evidence that he trusts anyone whom he has not known from the anti-Soviet jihad. In the 1990s, al Qaeda incorporated the brightest and most dedicated novices who came to train in its network of camps in Afghanistan. They became its cadres and trainers. In the past five years, al Qaeda has not been able for the most part to incorporate new recruits among its ranks. Western novices traveling to Pakistan in the hope of making contact with al Qaeda have been turned around and sent back to the West to carry out terrorist operations. Meanwhile, the success of the predator drone strike campaign on the Pakistani border has dramatically thinned the ranks of both al Qaeda leaders and cadres. Now it appears that these strikes are also targeting al Qaeda allies with a transnational agenda.
7. Protection of Western homeland involves an effective strategy of containment of the threat in the Afghan Pakistan area until it disappears for internal reasons. In the past five years, al Qaeda or its transnational allies have not been able to infiltrate professional terrorists into the West, as Ramzi Youself did in New York in 1993 or the GIA did in France in 1995. None of the plots during that time involved any full time professional terrorist. This is probably due to good cooperation among intelligence agencies around the world, good intelligence databases and increased vigilance and security at airports around the world. To carry out operations in the West, these global neo-jihadi terrorist organizations are completely dependent on Western volunteers coming to the Pakistani border to meet terrorist groups or on inspiring young Western terrorist wannabes to carry out operations on their own without any guidance or training. These organizations are stuck with the people traveling to the border area to meet with them, mostly through chance encounters. These travelers are relatively few in number, totaling in the dozens at most. The emerging details from the terrorist trials and the interrogations of the Westerners captured in Pakistan are quite clear on this score. Terrorist organizations can no longer cherry pick the best candidates as they did in the 1990s. There is no al Qaeda recruitment program: al Qaeda and its allies are totally dependent on self selected volunteers, who come to Pakistan. Global neo-jiahdi terrorism also has no control over the young people who wish to carry out operations in the West in its names. The result is a dramatic degradation of the caliber of terrorist wannabes, resulting in the decrease in success of terrorist operations in the West despite the increased number of attempts. Containing those who travel to Pakistan for terrorist training is a counter-terrorism problem and is much easier problem to solve than transforming an adjacent nation through a national counter-insurgency strategy. The West has been doing well in this strategy of containment with Pakistan‟s active collaboration.
8. The decrease of global neo-jihadi terrorism in the last five years is testimony to the effectiveness of international and domestic intelligence as well as good police work. The timeline analysis of global neo-jihadi terrorism shows that the major threat to Western homelands is al Qaeda inspired homegrown networks. Disrupting such homegrown plots has always been a domestic counter-terrorism mission through domestic intelligence and law enforcement. Indeed, there is a strong probability that the proposed counter-insurgency military surge may result in moral outrage in young Muslims in the West, who would take it upon themselves to carry out terrorist operations at home in response to the surge – just as the invasion in Iraq resulted in a dramatic increase in terrorist operations in the West. So, far from protecting the homeland, the surge may actually endanger it in the short term. After going through a learning process, Western law enforcement agencies, in coordination with their foreign counterparts, have done an effective job in protecting the homeland.
9. In conclusion, counter-terrorism works and is doing well against the global neo-jihadi terrorist threat. It consists of a combination of good domestic police work, good domestic intelligence, good cooperation with foreign domestic intelligence agencies, good airport security, good border control, keeping up the pressure on al Qaeda and its transnational allies in Pakistan through arrests and Predator drone attacks, using political and economic skill to deny terrorist sanctuary in Pakistan, supporting the Pakistan military to dislodge foreign militants from Waziristan while sealing the border on the Afghan side, and continued sanctuary denial in Afghanistan. These are measures that will continue regardless of what is done in Afghanistan. There is definitely no necessity and very little value added for the counter-insurgency option, which is the most costly in terms of blood and treasure, probably the least likely to succeed and may even make things worse in the short run.
10. Counter-insurgency and nation building in Afghanistan may be important for regional reasons and I would be honored to address these complex issues at another time. I am pleased to see that the committee invited my former chief, Milton Bearden, to testify on these issues last week. I had the privilege to serve under him in Islamabad, where I spent almost three years in personal contact with the major Afghan Mujahedin commanders fighting against the Soviet Union. I stand ready to comment on counter-insurgency strategy and tactics in Afghanistan based on my personal experiences with important Afghan insurgents. But counter-insurgency in Afghanistan has little to do with global neo-jihadi terrorism and protecting the homeland.

Notes

1. For an analysis of the concept of jihad through Muslim jurisprudence and history,see Richard Bonney, 2004, Jihad: From Qur’an to bin Laden, New York: Palgrave Macmillan↩
2. See Lawrence, Bruce, ed., 2005, Messages to the World: The Statements of Osama bin Laden, London: Verso, page 61↩
3. See Sageman, Marc, 2008, Leaderless Jihad: Terror Networks in the Twenty-First Century, Philadelphia: University of Pennsylvania Press↩
4. See Mountstuart Elphinstone, 1815, An Account of the Kingdom of Caubul, and its Dependencies in Persia, Tartary, and India; Comprising a View of the Afghaun Nation, and a History of the Dooraunee Monarchy, London: Longman, Hurst, Rees, Orme, and Brown; Olaf Caroe, , 1958, The Pathans: 550 B.C. – A.D. 1957, Oxford: Oxford University Press; and Akbar Ahmed, 2004, Resistance and Control in Pakistan: Revised Edition, London: Routledge, for sophisticated examples of effective governance in the Afghan tribal areas.↩