Very interesting stuff; clearly I need to become better versed in the specifics of the "organized" vs "disorganized" argument that's making the rounds. I tend to come at this from a specific perspective; in consequence, I'm not sure that my gut instinct about how to interpret recent developments is not all that different from what the "organized" position is claiming. Put differently, these positions are less exclusive than some might assume.

A quick point: if one is chiefly concerned with single incident of high-value attacks, then an organized network is a very bad thing. If one is concerned with the long-term durability of that network and to its ability to put constant geostrategic pressure on US interests, there are reasons to fear a more diffuse movement. As with any network structure, there are pros and cons associated with a more centralized or a more decentralized pattern.

Glad you enjoyed the post, I'd recommend Rohan Gunaratna's Inside Al-Qaeda and Jason Burke's Al-Qaeda: Casting A Shadow Of Terror if you want to fairly decent representations of the two schools of analysis. I agree with you that the distinctions are not as exclusive as some like to make out and would even go as far as saying that al-Qaeda may exhibit the characteristics of both schools of thought periodically depending on the situation.

Agreed with your point at the end, though I would note that the death toll from a sustained terrorist campaign is likely to result in far less casualties long-term than one that involves a successful mass casualty attack. 9/11, for instance, easily trumped the entire Intifada in terms of body count and if we are to take the Jordanian government at its word the chemical plot Zarqawi wanted to launch against Amman would have resulted in more net deaths than the whole of Operation Iraqi Freedom and the insurgency depending on which Iraqi casualty count one subscribes to. A couple of nukes detonated in major cities certainly would.

A very fine job as usual, Dan.

Mr. Darling, I largely agree with with your arguments, and those of Mr. Gunaratna. However, I feel there's something missing here. Mr. Nexon is surely correct when he says, "Put differently, these positions are less exclusive than some might assume." Sometimes Al Qaeda (or IIF - International Islamic Front as the Indians call it, which I believe is really a better term, but Al Qaeda is the popular term...) appears to be well-regimented, with a distinct command pathways, and other times it appears quite limp.

It is as though we are studying an animal that seems to change phylum, depending on which behavior we observe. Sometimes this animal is a chordate, sometimes not. Why does al Qaeda appear this way?

In the main, the behaviors we observe from al Qaeda come from its offensive actions. This is when we see al Qaeda revealed - both in the commission of these acts, and in the pursuit by authorities. There are occasional things we learn about al Qaeda from attacks on it - the Golden Chain list came from a raid on a "charity" office in Bosnia. But we learn the most about Al Qaeda when bombs go off.

I submit (and I think you'll mostly agree) that not all of the attacks we see come from the same "root command" shall we say. It's difficult to find proper language to describe this... Although I believe al Qaeda maintains a central command and control, not all attacks that al Qaeda carries out directly serve the core goals of the organization.

I beleive there are three different types of attack, each type characterized by the level of central organization behind it.

1. An attack launched by an al-Qaeda affiliate. This type of attack is launched without the direct authorization or maybe even knowledge of al Qaeda's leadership. This attack is launched for local reasons; fundraising, e.g. an Abu Sayyaf kidnapping; propaganda/psywar, e.g. assassination of van Gogh; target of opportunity in active combat zone, e.g. IED or sniper attack Iraq, Chechnya.

This kind of attack, though horrible in its own way, does not present a challenge to the existence of a state. In most cases, these affiliate organizations by themselves are no more than an irritant to a government.

2. An attack launched by an al-Qaeda affiliate or cell with affiliate support. This type of attack is launched on the explicit orders of the central leadership. This attack is launched to advance the cause of al Qaeda's central leadership i.e. re-establishing the Caliphate and expanding Islam into the Dar al-Harb. A good example of this would be the 2002 Saudi attacks cited above.

An attack of this type will be much more serious than the first type of attack, especially if the security organiztions of the target country do not have the situation in hand shortly after. Such an attack may well be dangerous to a government's survival where large sections of the population are already oppposed to the government and in favor of Al Qaeda. I'm thinking in particular Saudi Arabia and Pakistan.

I believe that Al Qaeda would like a central headquarters again, instead of spreading out training camps and command shuras piecemeal. Such a headquarters country would not only ease operations, but also start the road to re-establishing the Caliphate. In Sudan, and then Afghanistan, bin Laden made himself and his organiaztion indespensible to the rulers of those lands. This provided a degree of security for Al Qaeda, and also put some of the levers of state power in Al Qaeda's hands - coining money, making passports, diplomatic pouches. Bin Laden would never become the de jure leader of any state, but his aim may be de facto leader. The terrorists eat the state from within the government, rather than the reverse.

Turabi may have seen the danger, and the US ultimatum provided the excuse he needed to show bin Laden the door. When bin Laden became Mullah Omar's right-hand man, bin Laden found the benefits of coopting the Taliban government were not as fruitful. That government wasn't widely recognized, and Omar didn't show much interest in events outside Afghanistan.

3. An attack launched by a state, using al Qaeda as a contractor. This type of attack is launched on the orders of a sponsor state. The central core makes its own cells, support organizations, and affilliates available for the operation. The state will augment al Qaeda's command and control with its own. The state will also supplement al Qaeda with some of the state's own covert resources for the operation.

This type of attack is the most highly organized, and the most perilous. It can topple governments and pose a danger to entire societies. As Mr. Nexon writes, "...if one is chiefly concerned with single incident of high-value attacks, then an organized network is a very bad thing."

The true high-value attacks are indeed backed up by an organized network, but it is a network that is not wholly organic to Al Qaeda. Al Qaeda "purists" (as I call them) like Mr. Bergen or Mr. Gunaratna see only bin Laden and his band of Merrie Men at the heart of each bombing. Ms. Mylroie sees the hand of Iraq (or Iraqi Former Regime Elements) in every attack. Mr. Ledeen sees Iran's nefarious finger pushing the buttons. They're all wrong and they're all right. Bin Laden wants to persue his own goals -- his Golden Chain backers contribute to advance his agenda. But he's not strong enough to do it on his own. So he must cooperate with states. Spectacular, one-time attacks are costly. None more so than 9/11, but at least he's got a fair bit of street cred out of it.

Rory,

Excellent comments. Here's the issue in abstract terms. Let's assume that the al-Q movement has some sort of a core-periphery structure (C = core leadership, P = cells, and O = omegas, i.e., sympathizers, potential recruits, walk-ins):

O-P=C=P-O

(I am not saying this is the structure of the movement, just using it as a hypothetical)

The "organized" vs. "loose" debate that we're having here pivots on the degree of hierarchy found in the P=C=P relationship. In this representation, we're talking about the existence of ties - P=C=P are connected to one another. If the relationship were highly centralized, we'd expect PP, with C exercising clear hierarchical/operational control over P.

Now it may be that we are not looking at a strict hierarchy but we are seeing the impact of (1) C (core al-Q) being central to the structure, (2) thus being a key broker in the movement, and (3) therefore being able to exercise a strong influence over the activities of P1... Pn. If that's the case, then we would expect al-Q to be at times be "well-regimented" but at other times appear "quite limp."

Just to chime in briefly on one point. While I agree that a mass-casualty attack requires leadership and coordination (different from a less hands-off command and control) which is less desirable than a shallower and less-capable diffuse-yet-degraded network, one thing to keep in mind: fragmentation makes groups/movements more durable -- thus, while the short-term threat is degraded, the long-term threat is that a surviving element(s) that had hived off of the original organization one day generates the capacity for a mass-casualty attack. Sorry for the run-on sentence, just one more analytic point. Interesting discussion.

Mr Nexon,

Any diagram or flow-chart of al Qaeda would certainly be quite convoluted!

I think that "Omegas" sympathizers, potential recruits, etc. are first absorbed into associate organizations, and are invited to join the central core only after an appropriately impressive time there. For an example of this, check out My Jihad by Aukai Collins.

My proposed diagram (within the limits of this medium of course) would also include a "G" for Group - one of the terrorist organizations or guerilla groups. These would be autonomous groups independent of the cells that we hear so much about in our ever-informative news media.

G=G=C=G=G might be a 'normal' state of affairs for al Qaeda. But this may be slightly deceptive. Terrorist and guerilla operations are expensive. Money flows out from the core, from the financial comittee. (Mr. Darling knows much more than I do about that.)

The central core, I believe, controls the cells directly. PP These include sleeper cells, planted by the core for long-term goals, and sent out to commit specific actions. An example of one of these actions might be the Keynan hotel bombings/ attempted airliner shootdown. When the core wishes to influence a group, a cell sent from the central core is appended to a group.

G=C=G+P

Interesting comments and analysis by all.

One of the main points that needs to be looked at is just how decentralized an organization has to be in order to classify it as a movement and vice versa.

I agree that extreme decentralization is more effective to the long-term survival of any terrorist organization, but consider that extreme decentralization is basically what Hamas has been forced to implement in light of the Israeli penchant for assassinating its leaders and the result (thankfully) is that very few of the attacks planned by Hamas cells are not actually carried out because they keep having to reinvent the wheel.

Al-Qaeda strikes me as being more interested in destruction than durability because they seem to believe (correctly, in my view) that the more destructive they are against the West the more enduring their movement will be because the more it looks to the average Muslim like they're going to win out in the end. Like I said earlier, they easily trumped the entire Intifada with a single attack on 9/11 at the (initial) cost of 19 operatives and about $300,000, so I wouldn't say that a more centralized approach doesn't have its own advantages to the situation.

I think one thing that keep people leaning toward the 'supreme decentralization' side of this is the "vanguard theory" notion applied to al-Qaeda. Decentralization -- in this sense -- is a core component of its agenda; to evolve into a movement. I don't think it's catalyzed a mass movement by any means, but does it have to? I'm not sure. At any rate, I'm increasingly persuaded by Dan's take.

I was reading an interesting book that was first published in 1990 and republished in 1998 called "Origins of Terrorism: Psychology, Ideologies, Theologies and States of Mind."

One thing that this book deals with is how terrorist organizations and "cells" are created, how they "recruit". A core theme in the book is that the members of these organizations are not necessarily recruited from a top down directive and list of requirements for recruits, but a bottom up, often "self selection" process in which the recruits/originating members of a cell determine that they a) are angry at a collective idea; b) cannot obtain satisfaction from normal processes (ie, like politics, protesting,etc) and c) determine that the only action that will get notice is violence.

For individuals, this often results in a "search" for something larger to belong to and it results in a psychological process that allows them to view their crimes, as heinous as they are, as "moral" when taken for the "greater cause".

I believe the issue of decentralization is really about how recruits or self selecting cells are found or self select to join the cause and is confused with the general organization structure. The second thing that appears as "disorganized" or "decentralized" is that, when attacks occur, it seems different groups, if not "new" groups with different names pop up to claim the attacks.

I recall reading interrogations of the Guantanamo prisoners and, when asked about "Al Qaida" many said they had never heard of that group. Some people think that this was a ploy or simple denial to protect themselves. Others think that it is because a central "Al Qaida" doesn't exist.

I beleive the reality was stated by an FBI officer who assisted in the interviews indicated the problem was more about what we called the group as opposed to what they called themselves. It is also in the way that these groups set up operations in Afghanistan and Pakistan. Many groups called themselves by different names and recruited from different regions or different "teachings" (this goes along with the concept that Islam and even Islamist doctrine comes from more than just the Wahhabi sect and that there are more than two ideas of Islam floating around). They each set up their own "safe houses" and had religious teachers administering to the groups.

However, Zawahiri himself wrote in his book, Knights Under the Prophets Banner, that he took some time to get all of these groups together, bringing the leaders to bin Laden and basically giving them all a joint and centralized idea of what Islamism was and should be, developing cooperation, centralized organization and strategic goals.
These were eventually added to the roster that bin Laden was famous for starting (Holy War, Inc) and called it "The Base" or "Al Qaida" as we became familiar with it.

These leaders act more like the joint chiefs of staff for our own armed forces, planning operations, getting funds, mobilizing forces, etc. However, like in all wars, there is a central idea and strategy. This comes from bin Laden and Zawahiri and was made plain in their statement of war against Crusaders and Jews. The others under them simply act like our military apparatus.

Think of it as the multiple arms of the military. They have a core plan and directive, a strategic goal, but when in the field, the individual units plan and execute operations within the realm of the original strategy laid out by the command structure and sometimes these are single unit activities and sometimes they involve "joint arms" operations.

Just like in our own command structure, units do not report and ask for permission from the president, the JCOS or SecDef for every mission. Each unit knows the over all strategic goals and acts on it. The only time it goes up the chain of command to the Sec Def, JCOS or President is when those operations might change or widen the over all strategic objectives.

It seems clear that this is how "Al Qaida" or whatever they refer to themselves, operate. The "new names" or different names of groups claiming responsibility are simply "unit names" and they serve a secondary purpose, because their organization is not clear to us, it makes it seem that there are many more people involved and many more people joining the "cause" and serves to keep their organizational structure "in the fog" for us.

Aside from the over all strategic goals outlined in the declaration of war, other messages and operations point to centralized command structure.

First is that the attacks from 1993 WTC to 9/11 were done through a central command, funding and strategic planning structure. It doesn't stand to reason that an organization that thrived on this central command structure would change its over all operation activity simply because they lost their original base. There is a matter of "comfort".

Secondly, Zarqawi organizes in Iraq and then AQ "blesses" his organization as a member of Al Qaida. This is more like a commander raising a unit and then receiving its unit designation and "flag". Zarqawi then sends letters about the situation in Iraq. Osama gives directives through video and audio messages. He then directs Zarqawi to do more operations against the "west". shortly after, we see attacks in England. Central command and control? Or somebody making a "suggestion" as in the old stories about Mohammed, and then taking it on themselves to "please the prophet"?

The nature of the attacks in Europe, from 3/11 to 7/7, also show an overall strategic planning and goals.

Each of the attacks were aimed at specific infrastructure and had multiple strategic goals. The attacks were planned against structures that have a financial impact, impacted transportation, would cause many civilian deaths and injuries(when executed properly). The attacks took place during political unrest and when political change was most likely (that includes the London bombing when G8 was there and when the Labour Party won with a reduced majority; there were questions about whether Blair would be replaced by Brown).

All of this points to superior strategic thinking and centralized command and control, regardless of how the "field units" are pulled together or recruited.

I was reading an interesting book that was first published in 1990 and republished in 1998 called "Origins of Terrorism: Psychology, Ideologies, Theologies and States of Mind."

One thing that this book deals with is how terrorist organizations and "cells" are created, how they "recruit". A core theme in the book is that the members of these organizations are not necessarily recruited from a top down directive and list of requirements for recruits, but a bottom up, often "self selection" process in which the recruits/originating members of a cell determine that they a) are angry at a collective idea; b) cannot obtain satisfaction from normal processes (ie, like politics, protesting,etc) and c) determine that the only action that will get notice is violence.

For individuals, this often results in a "search" for something larger to belong to and it results in a psychological process that allows them to view their crimes, as heinous as they are, as "moral" when taken for the "greater cause".

I believe the issue of decentralization is really about how recruits or self selecting cells are found or self select to join the cause and is confused with the general organization structure. The second thing that appears as "disorganized" or "decentralized" is that, when attacks occur, it seems different groups, if not "new" groups with different names pop up to claim the attacks.

I recall reading interrogations of the Guantanamo prisoners and, when asked about "Al Qaida" many said they had never heard of that group. Some people think that this was a ploy or simple denial to protect themselves. Others think that it is because a central "Al Qaida" doesn't exist.

I beleive the reality was stated by an FBI officer who assisted in the interviews indicated the problem was more about what we called the group as opposed to what they called themselves. It is also in the way that these groups set up operations in Afghanistan and Pakistan. Many groups called themselves by different names and recruited from different regions or different "teachings" (this goes along with the concept that Islam and even Islamist doctrine comes from more than just the Wahhabi sect and that there are more than two ideas of Islam floating around). They each set up their own "safe houses" and had religious teachers administering to the groups.