http://www.interesting-people.org/archi ... 00102.html
> In a Single Night
> By EDWARD N. LUTTWAK
> Many commentators argue that a pre-emptive air attack against
> Iran's nuclear installations is unfeasible. It would not be swift
> or surgical, they say, because it would require thousands of strike
> and defense-suppression sorties. And it is likely to fail even then
> because some facilities might be too well hidden or too strongly
> protected. There may well be other, perfectly valid reasons to
> oppose an attack on Iran's nuclear sites. But let's not pretend
> that such an attack has no chance of success. In fact, the odds are
> rather good.
> The skeptics begin sensibly enough by rejecting any direct
> comparison with Israel's 1981 air attack that incapacitated the
> Osirak reactor, stopping Saddam Hussein's first try at producing
> plutonium bombs. Iran is evidently following a different and much
> larger-scale path to nuclear weapons, by the centrifuge
> "enrichment" of uranium hexafluoride gas to increase the proportion
> of fissile uranium 235. It requires a number of different plants
> operating in series to go from natural uranium to highly enriched
> uranium formed in the specific shapes needed to obtain an explosive
> chain reaction. Some of these plants, notably the Natanz centrifuge
> plant, are both very large and built below ground with thick
> overhead protection.
> It is at this point that the argument breaks down. Yes, Iraq's
> weapon program of 1981 was stopped by a single air strike carried
> out by less than a squadron of fighter-bombers because it was
> centered in a single large reactor building. Once it was destroyed,
> the mission was accomplished. To do the same to Iran's 100-odd
> facilities would require almost a hundred times as many sorties as
> the Israelis flew in 1981, which would strain even the U.S. Air
> Force. Some would even add many more sorties to carry out a
> preliminary suppression campaign against Iran's air defenses (a
> collection of inoperable anti-aircraft weapons and obsolete
> fighters with outdated missiles). But the claim that to stop Iran's
> program all of its nuclear sites must be destroyed is simply wrong.
> An air attack is not a Las Vegas demolitions contract, where
> nothing must be left but well-flattened ground for the new casino
> to be built. Iran might need 100 buildings in good working order to
> make its bomb, but it is enough to demolish a few critical
> installations to delay its program for years -- and perhaps longer
> because it would become harder or impossible for Iran to buy the
> materials it bought when its efforts were still secret. Some of
> these installations may be thickly protected against air attack,
> but it seems that their architecture has not kept up with the
> performance of the latest penetration bombs.
> Nor could destroyed items be easily replaced by domestic
> production. In spite of all the claims of technological self-
> sufficiency by its engineer-president, not even metal parts of any
> complexity can be successfully machined in Iran. More than 35% of
> Iran's gasoline must now be imported because the capacity of its
> foreign-built refineries cannot be expanded without components
> currently under U.S. embargo, and which the locals cannot copy.
> Aircraft regularly fall out of the sky because Iranians are unable
> to reverse-engineer spare parts.
> The bombing of Iran's nuclear installations may still be a bad idea
> for other reasons, but not because it would require a huge air
> offensive. On the contrary, it could all be done in a single night.
> One may hope that Iran's rulers will therefore accept a diplomatic
> solution rather than gamble all on wildly exaggerated calculations.
> Mr. Luttwak is a senior fellow at the Center for Strategic &
> International Studies.
Weblog at: <http://weblog.warpspeed.com>