The first impulse of any lawyer in coming at a question is to try to shift the burden of proof. So that's the first thing I'm going to try to do. I want to establish that the burden of persuasion is always on those who call the status quo "illegitimate". Here goes:
Legitimacy is the flipside of revolution. Clearly, if a revolution is justified, then the existing order must be illegitimate. On the other hand, if the status quo is legitimate, then revolution would be unacceptable. So I'll stipulate as a definition of "legitimacy" that quality that can be ascribed to a regime when revolution against it is wrong.
Revolution is like war in that its direct consequence is the foreseeable and typically intended death of human beings and its indirect consequences are typically far-reaching but unforeseeable. It follows, I think, that there should be a presumption against revolution, as against war. The absolute pacifist position -- that war and revolution are never justified -- may be utopian, but they obviously have greater appeal than the opposite position that we should be in perpetual civil and international conflict.
It follows, therefore, that there is a presumption against revolution or, equivalently, a presumption of legitimacy. As Thomas Jefferson, in a statement against interest, put it:
Prudence, indeed, will dictate that Governments long and established should not be changed for light and transient causes...
However, since governments are not divinely- but humanly- ordained, the presumption of legitimacy can be rebutted by tyranny, just as the presumption against war can be rebutted by aggression:
But when a long train of abuses and usurpations, pursuing invariably the same Object evinces a design to reduce them under absolute Despotism, it is their right, it is their duty, to throw off such Government, and to provide new Guards for their future security
Now, I share Yglesias' skepticism that the British Crown in the 18th century was in fact engaged in a long train of abuses and usurpations, with a design to reduce white Protestant American colonists under absolute Despotism. Au contraire. If you read the particulars appended later in the Declaration, they are either not very oppressive at all (like giving Quebec the right to use the Civil Code, or guaranteeing traditional Indian lands), grossly exaggerated ("He is at this time transporting large Armies of foreign Mercenaries to compleat the works of death, desolation and tyranny, already begun with circumstances of Cruelty & perfidy scarcely paralleled in the most barbarous ages") or pretty darn vague ("He has refused his Assent to Laws, the most wholesome and necessary for the public good").
But, merits of his case aside, I'm willing to grant that in this part of the Declaration, Jefferson gets the issue right. The question of legitimacy is a question of practical judgment: is the regime so bad, so thoroughly beyond reform, that revolution is a lesser evil?
A regime need not be particularly democratic to be legitimate. Indeed, its legitimacy may depend on what the consequences of revolution are likely to be: if particularly bad, then even a lousy regime may become legitimate. In fact, if you accept the prudential judgment of the ancients that a democracy unmitigated by elements of aristocracy is particularly likely to fall into tyranny, then a regime can be more legitimate when less democratic.
Using this yardstick, the legitimacy of judicial review seems like a more easily resolved problem. Judicial review only becomes illegitimate when it creates sufficiently great degree of tyranny to justify a revolution. Few of us think that judicial review in either Canada or the United States has got to that point. (Those who do are advised to confine themselves to peaceful propagation of their insane belief system.)
Have I cheated here? Do people who ask about the legitimacy of counter-majoritarian judicial review really want to know whether it justifies revolution?
Maybe not. Maybe they just want to discuss what (normatively) is the best approach the courts should take to judicial review. Or what role it should play in the ideal constitution. Both interesting questions, but not, as I've defined things anyway, questions about legitimacy. The US and Canadian Constitutions can be legitimate if non-ideal. However, this Tory point prevails: improvements that aren't possible within their structure are to be abandoned if they don't justify revolution. Improvements that are possible within their structure should be discussed as possible reforms, and their costs and benefits worked out.