Euvoluntary Rebellion?

Hume's Essays contain a treatise on the legitimacy of what he calls "passive obedience" (Part 2, essay XIII in the Liberty Fund edition). He, like Locke after him, argued that under normal circumstances, ordinary citizens owed fealty to the king for pretty much utilitarian reasons. In his words, "common sense teaches us, that, as government binds us to obedience only on account of its tendency to public utility, that duty must always, in extraordinary cases, when public ruin would evidently attend obedience, yield to the primary and original obligation." In this case, the primary and original obligation is "Salus populi suprema Lex", the supreme law of the land is the ~safety~ of the people. "Safety" is an imperfect translation of "salus" these days, but that might be more a transformation of vernacular English than a flaw in the original Hume.

Hume then goes on to say more or less that the public owes to itself an unflinching primary duty in accordance with this first Law to rebel in cases "when the public is in the highest danger, from violence and tyranny." Unsurprisingly, he nodded in reflective approval at England's shrugging off the reprehensible James II in the Glorious Revolution in 1688. This follows naturally from his comments on coercion. If you believe that voluntary exile tends to be spectacularly non-euvoluntary ("Can we seriously say, that a poor peasant or artizan has a free choice to leave his country, when he knows no foreign language or manners, and lives from day to day, by the small wages which he acquires?") and you believe that in the presence of tyranny, the power of petition holds no suasion over the sentiments of the prince, the lowest-cost alternative is to overthrow and depose the sovereign.

Common sense, right? Not so fast, says Bryan Caplan, who makes a set of compelling arguments in favor of peace, also grounded in ordinary pedestrian morality. In Caplan's world, war is hell (and Bryan has made no small effort to catalog the horrors of totalitarian regimes) and its benefits are murky. For open conflict to make sense, you'd need to apply rigorous cost-benefit analysis, the kind that ordinary members of the public are particularly bad at conducting (cite: most of the public choice literature, try Mueller's Public Choice III for a start, then Caplan's own Myth of the Rational Voter). Both Caplan and Hume counsel peace as the default, though one's is domestic, the other's, foreign. They differ in that Hume gives two criteria for legitimate rebellion:

The first is, that [warmongers'] antagonists carrying the doctrine of obedience to such an extravagant height, as not only never to mention the exceptions in extraordinary cases (which might, perhaps, be excusable) but even positively to exclude them; it became necessary to insist on these exceptions, and defend the rights of injured truth and liberty. The second, and, perhaps better reason, is founded on the nature of the BRITISH constitution and form of government.

He then goes upside the head of the divine right of kings (as he is so very wont to do), presaging the revolt in the colonies and Jefferson's language in the Declaration of Independence. Contrarily, Caplan would probably say, "you know, the gap between life under British rule and American sovereignty probably wasn't bad enough to justify either the immediate loss of life in the Revolution or the ex ante relief of George III's "onerous" taxation (try not to giggle if you can) or his actual star-chamber, general-warrant despotism.

Hume and Caplan both seem to suggest a hierarchy of behavior.

1. In ordinary circumstances, obey the sovereign will
2. If things get bad, pack up and head for greener pastures
3. If things get really bad and your back is up against the wall, pick up a rifle

(2) and (3) are Scylla and Charybdis if we're to take both propositions seriously (and I see no reason not to). A reasonable exercise is then to consider the relevant margins of (in order) voice, exit, and rebellion and what a foreign polity might do to provide aid and comfort to the subjects of tyranny.

From abroad, voice is diplomacy (including foreign aid). This we already use for better or worse, and to the extent that there's a non-null core to the game, savvy diplomats can, by hook or by crook, avert insurrection. Bravo for those instances. But the core is occasionally null, as it seems in at least a few Middle Eastern countries over the past several decades. 

So if voice fails, what is the just response when a rebellion is underway in foreign lands? Hume did argue against relocating to bizarre alien lands, but I'm sure he'd agree that making a pittance is preferable to ending up on the business end of a spear. And Caplan is one of the great modern champions of open borders. Would not both these fine gentlemen argue in favor of nations eagerly accepting refugees? We might reasonably argue about the basic functions of government, but surely among the list would fall fundamental protection of life, right? Shouldn't any decent Rawlsian or Hobbesian or Lockean or Caplanian support at a bare minimum open borders for war or rebellion refugees?

Why does it seem then that the commonplace modern reaction to tyranny abroad is to impose economic sanctions, to lob ordinance, or even more flaccidly, to send UN inspectors? I can understand how elites would want to wage war (hint: it burnishes their legacy) but the public's complicity baffles me. It seems as if outright blood-in-the-streets rebellion is an execrable outcome , far worse than the slights that motivate typical swells of humanitarian sentiment. What's the difference? What am I missing? Is rebellion actually euvoluntary and I'm just not seeing it? What gives?