During the carnage in Lebanon, where in numerous cases, noncombatants were killed by precision weapons that missed their targets, and where in response, relatively few Israeli noncombatants were killed in spite of the rockets fired against them being extremely imprecise, I sought to understand what seemed to be a paradox: why were the very precise Israeli weapons killing more civilians than the less precise Hezbollah weapons?
Some have used this paradox to strengthen their belief that Israel is deliberately targeting civilians. I would prefer not to believe this, although it is a proposition that is difficult to disprove. Others have pointed to the fact that Hezbollah deliberately placed strategically important assets close to civilians, although why the great disparity in the kill rate between the imprecise Hezbollah rockets and the precise Israeli ones?
Well, I've come up with a statistical explanation that may account for part of the paradox.
It is necessary to assume, for the purposes of this explanation, that Hezbollah indeeed places its assets among civilian residences, because it understands the value of their deaths for propaganda purposes. It is also necessary to assume, again, just for the purposes of this account, that Israel uses its most precise weapons against targets known to be close to civilians.
Now, precise weapons, whether Israeli or American, are not precise in any magical way. What greater precision means in this context is that there is a much greater probability that the bomb or missile will strike close to its intended target. There are several sources of error here: someone has to locate the target, someone has to program the device, and then there are various possible errors in the actual trajectory of the missile or bomb. All of these errors, taken together, cause the distribution of distances from the actual target to the place where the missile strikes to have a small variance, compared to less accurate missiles. If you created a composite scatter plot in which the intended target of all of the Israeli missiles in this offensive were superimposed, you would see a very dense region on and very near the target, with gradually less dense regions as you move away from the target. I think this concept is fairly clear.
However, you will see relatively few of the missiles exactly on the target, where "target" is defined as the precise blast-sized region centered on the exact coordinates of the Hezbollah resource. Instead, you will see that a large percentage, possibly a majority, of the missiles will instead land close to the actual target, often close enough to damage it, but not always.
If you compare this kind of scatter plot to one for the Hezbollah missiles being fired at Israel, you would see something very different. In this case, almost none of the missiles would be close enough to its target to damage it; the probability of being "close" to the target is, apparently, nil. They will obviously hit things and damage them, but the scatter plot of their closeness to the target will be much more spread out.
Now go back to the strategy on the part of Hezbollah of placing their assets close to civilians, in the hope of scoring propaganda points. Remember that most of the precision weapons used by the Israelis land close to, not directly on, their targets. And what you get is a very predictable kill zone in the region right around military assets. This prediction has been confirmed now quite a few times during the Israeli bombing of Lebanon.
Ironically, if Israel had wanted only to fire the same number of missiles but minimize the number of civilian deaths, then they should have used Hezbollah-style, inaccurate missiles. The use of precision weapons, combined with a "human shields" strategy of asset placement, will alway result in a large number of civilian deaths.
Combine this statistical prediction with the large number of inaccurate missiles being fired by Hezbollah, resulting is annoyance, random property damage, and fear, but in relatively less civilian casualties, and you have a probaganda bonanza for Hezbollah, and the corresponding PR disaster for Israel.
What blows my mind is that the Israelis hadn't figured this out in advance. Now, in hindsight, and with the statistical model in mind, it seems totally obvious.
Greg Shenaut