I'm not sure what submarines have to do with human nature, but, in any case, William Saletan has devoted a column to discussion of the threat posed to the U.S. by submarine-operating terrorists. Saletan points out that drug cartels have used submersibles to good effect -- though the amount's debatable, he says they transport up to a third of the U.S. cocaine supply. Cartels value the subs because they're difficult to detect, and because they have heretofore enjoyed legal near-immunity. When a drug-carrying submarine is identified by the Coast Guard or the Navy, the crew abandons ship and scuttles the boat, sending the evidence to the bottom. Laws regarding operation of submersibles have changed, limiting this legal immunity, but this mode of conveyance remains attractive to drug dealers. Compared to the price of the cocaine they carry, submarines are relatively inexpensive; the models currently in use are thought to cost about $1 million.
Saletan then jumps from drug cartels to terrorists. If drug dealers can use submarines to good effect, he suggests, why can't terrorists? Terrorists might use submarines to deliver some kind of radioactive payload to a US port, even (although Saletan doesn't specifically mention this) a nuclear weapon. While I'll concede that aquatic terrorists do pose some threat, Saletan runs off the rails by failing to consider the practical difficulties that such an operation would face. The chokepoint for terrorists isn't really acquiring the submarine; I suspect that one could be bought, and I also suspect that if submarines haven't already been used to smuggle goods into Gaza, they soon will be. The problem is finding a payload that would a) not kill the crew before it reached its destination, and b) be worth delivering. A radioactive "dirty bomb" even of significant size would have only minimal effect if detonated underwater; the whole point of the operation is to cover the target with radioactive material. More to the point, dirty bombs have thus far remained wholly in the realm of imagination, and I doubt it's because al-Qaeda hasn't been introduced to the submarine. Rather than thinking about new ways that such a device might be delivered, it might behoove us to think about why no such weapon has EVER been used.
To the extent that al-Qaeda submarines might pose a threat, the concern would be for a Cole-style attack against a U.S. naval vessel. A submarine laden with explosives instead of cocaine could do severe damage to a Navy ship, possibly even sending it to the bottom. Even in this case, however, the cost to al-Qaeda would be substantial; the submarine and the expertise needed to operate it would cost far more than the Cole attack, and would require a large network in whatever port area they used as a base. The upshot, as always, is that terrorism is generally far more difficult to conduct than it is to imagine.