I think his response conflates a question of efficacy (is it a good idea?) and probability (would they do it?). I tend to agree that a strike is probably on balance a bad idea for many of the reasons highlighted in Larison’s post.
But I also think that when push comes to shove Israel is willing to tolerate the risks associated with a strike much more than they are willing to tolerate the risks (as they see them) of not attacking.
Scoblete is right that I am basing my judgment about the likelihood of an Israeli attack on how likely it is that such an attack would achieve its objectives. Since it seems clear from all accounts that the Israeli military understands that an attack would not significantly delay Iran’s nuclear program, and they see that an attack would invite serious retaliation from Iran and its proxies that would threaten the civilian population in Israel, we have to believe that the Israeli government is willing to court immediate risks to its people for the sake of protecting itself against a threat that does not yet exist. The last two Israeli governments have made some blunders in recent years, but they cannot be so reckless as to start a regional war that wouldn’t even eliminate Iran’s nuclear program.