The eagle-eyed amongst you will have noticed that I am currently reading Michael Walzer's 'Just and Unjust Wars'. It was in part originally prompted by Walzer's desire to give a systematic account of the moral considerations which grounded his, and others', objections to the war in Vietnam. Walzer also supported the Israeli invasion of southern Lebanon, and, with some caveats, the bombing of Lebanese infrastructure, last summer. Although I haven't read the TNR article he wrote at the time - it's paywalled - I understand that, in part, his defence rested on a distinction between jus ad bellum and jus in bello (see here and here for a criticism and response in last winter's Dissent). His position seems to have been that whilst some Israeli acts were war crimes, the war itself was just. Whilst he does in general affirm the distinction between jus ad bellum and jus in bello in 'Just and Unjust Wars', I think it is worth recalling exactly what Walzer said about that distinction in the context of the war in Vietnam, and counter-insurgency campaigns in general. Although the parallel is not exact - Hizbullah was by no means solely conducting a guerrilla campaign against an Israeli invasion, and a significant part of Walzer's point is that the guerrillas are now the legitimate political authority, which means that making war on them is an act of aggression, a claim and entailment we would rightly hesitate to make about Hizbullah - I think it is nonetheless instructive.
In the theory of war, as we have seen, considerations of jus ad bellum and jus in bello are logically independent... But [once the guerrilla movement has won very substanial popular support] they come together. The war cannot be won, and it should not be won. It cannot be won, because the only available strategy involves a war against cvilians; and it should not be won, because the degree of civilian support that rules out alternative strategies also makes the guerrillas the legitimate rulers of the country... The position of the anti-guerrilla forces has become doubly untenable.
Even if the Israeli position was only singly untenable - because attacking Hizbullah was not, as such, an act of aggression, since Hizbullah were not the legitimate political authority - it was still untenable on this account, since Walzer's position is clearly that fighting a force which shelters amongst, and cannot be isolated from, because of their support for it, civilians will mean doing things that any justified rules of war will rule out. If ought implies can, and can has normative weight in the sense Walzer uses it in the above passage, then a war that cannot be fought should not be fought. Maybe Walzer would, because the campaign in Southern Lebanon had ostensibly more limited aims than the one in Vietnam - initially to return the kidnapped Israeli soldiers, and then to remove Hizbullah's capacity to launch rocket attacks into Israel - draw a distinction between the measures necessary to fight the kind of war being fought in Vietnam and that being fought in Lebanaon. This look spurious to me, though, because only the second of those ostensible aims justifies the extent of the Israeli incursion, and it is one which is functionally identical to the American aim in Vietnam; to destroy a guerrilla movement, since Israeli security against rocket attacks from a hostile Hizbullah could only ever be guaranteed by destroying it as an organisation. I made this point at the time, but it looks like Walzer as a matter of consistency ought to endorse it: sometimes the omelette could just never be worth the eggs.