There are many possible responses to the terrorist attack on the world trade center. Many people have called for an all out war on "terrorism". However there is much need for caution and it is unclear that enough caution will be taken.

The principal difficulty is that the US and NATO, are ready and armed to fight a physical war while the terrorists are fighting a psychological war. In order to "win" the war on terrorism in a convincing way, the psychological war must be won. It is difficult to win a psychological war with a physical war and there is a very significant danger that the physical war will actually increase the severity and frequency of terrorism in the long run. For example, even if Osama bin Laden is killed, attacking Afghanistan could easily result in even more people becoming willing to commit acts of destruction.

The word "terrorist" is too overloaded in order to continue using it in this essay. Instead, I will refer to "destructive acts" and "people who commit acts of destruction". This is done to clarify the subject of discussion.

The principle goal in a psychological war is to demotivate. In order to demotivate we must first define who we wish to demotivate. There are three groups in this world - people who would never commit destructive acts, people who are already committed to destructive acts, and the people who might or might not commit to destructive acts. This last group is the group whose opinion must be swayed in order to win any "war" on terrorism. These people must be convinced that destruction is not the right path and to isolate those committed to destructive acts. Our efforts should be bent towards this task.

There are both offensive and defensive ways to demotivate the undecided people. The notion of "offense" and "defense" are a bit loose in psychological war, so I will give them simple definitions: "offense" means things done external to this country while "defense" means things done inside this country. Both of offensive and defensive techniques should be employed to maximize effectiveness.

The US is currently very ready to go on the offensive. Going on the offensive can be effective, but there are several necessary caveats for offensive for a succesful offensive psychological war. First, targeting must be exceptionally good. Many of the techniques used in the past such as the embargo and bombing which occasionally struck civilians are as likely to motivate the undecided people towards destruction as away from destruction. It is critically important that public condemnation of those targetted be as strong and unanimous as possible. One standard technique for building unanimous support is via trial which should be considered here. In any case, the US must present convincing evidence motivating any target if it wishes to win the psychological war.

There is another form of offensive action which is far less likely to backfire - aid. After world war II, Germany was transformed from enemy to strong ally by the simple expedient of reconstructive aid. This was not necessarily an act of good will since the US did it partly to prepare for the cold war, but it was very effective, and it could easily be effective here. The principle difficulty with this approach is that several of the governments involved are corrupt and so funnelling aid to that nation may or may not effectively reach the people. However, in situation where we can trust the government and especially after any conquering, we should imply this technique. The resources invested in aid should probably be at least as much as the resources invested in direct military action since the aid approach has very little chance of backfiring. At worst, with a corrupt government, it will simply not be very effective.

One offensive technique which will not work in the long run is "intelligence". Intelligence is often valuable and effective at discovering attacks before they occur. However, current technological trends suggest that the ability to intercept and understand messages quickly and efficiently enough to discover attacks in advance will decline. In particular, it is now possible for anyone with an internet connection to communicate with anyone else with a significant amount of anonymity in a small amount of time via techniques which present any other party from understanding the content of the message. Ignoring these trends will simply give us a false sense of security which will be unexpectedly breached at some point in the future. Thus "intelligence" is not a solution, it is just a (sometimes inherently fallible) warning system.

The previous two techniques are offensive. Another significant set of techniques are defensive: make the US infrastructure a less attractive target. There are several ways in which this can be done.

One obvious technique is hardening the people of the US. It should not be the case that a random group of 100 people can be dominated by 4 or 5 men with knives. The US is already considering using 'Air Marshalls' which are armed under cover agents on every domestic flight. In addition to this, we should consider setting up a 'voluntary air marshall' program. Volunteers can submit to the training and testing necessary for an air marshall and then recieve the same duty/privilege of carrying a weapon on the plane and elsewhere. It is clear that solid standards for training and testing must be put in place and the credentials of any air marshall must be very difficult to forge.

It is important that the idea of a 'voluntary air marshall' extend well beyond the idea of planes. There are many other locations where destructive acts might be committed and the likelihood of such acts increases when we defend our commercial planes. Concerts, sports events, and any other large gathering of unarmed people might be a target.

There will be a cost to this approach. No acceptable system of training and tests will be foolproof. However it is difficult to argue that the failures of this approach will be as severely destructive as the world trade center destruction.

A second important defensive measure is dispersal. Technology is giving us the tools to avoid the need to travel (and be on planes) via teleconferencing. We can avoid the need to commute to large skyscrapers via telecommuting. The level of detail offered by news coverage of sports events is far higher than for most of those in the stands. The need for cities is no longer as important for an efficient economy.

There are also probably minor improvements in construction which would be effective. While it's unclear that skyscrapers could be hardened against fully fueled jetliners, stronger building codes might help with occasional bombs, biological warfare, and other destructive acts.

None of the defensive techniques is a silver bullet, but taken together they make the US into a substantially less attractive target. A somewhat more dispersed society with a nontrivial portion acting as armed guardians reduces the motivation for destruction. Acts of destruction will have reduced scale and effectiveness while the danger of failure increases.