There is no easy answer. To a certain extent, anyone who believes in the consent of the governed must concede that it counts in favor of Putin and his policies. But raw consent does not settle the matter, or we;' have to concede that Hitler and Stalin, both of whom were very popular while they were winning, were legitimate too. I underline that I do not put Putin in their league.
The point is that mere consent only goes so far. At the same time we must be very cautious about discounting the popular will. However, unless we are willing to give up radical projects, as Chris and Yoshie are willing to do for Russia and Iran but not, it seems, for the US,w here there ideas have even less purchase than they do in Iran or Russia, we cannot say that consent trumps.
[WS:] I do not want to sound overly simplistic, but did not Aristotle sort those things out? He classified government types by the number of rulers (one, few, many) and its functionally (correct and deviant), resulting in six types: 3 correct types (kingship, aristocracy, and polity), and 3 deviant types (tyranny, oligarchy, democracy). http://plato.stanford.edu/entries/aristotle-politics/#ConCit
Furthermore, the distinction between correct and deviant forms is NOT based on the consent of the governed, as Aristotle was keenly aware that opinions are not equal, but the system's capacity to realize the full human potential in its citizens (which is more objective than subjective opinions.) The latter is essentially a view espoused by Karl Marx as well.
What is important here is that consent of the governed (and the implicit legal positivism) offers no guarantee against a tyranny (cf. slavery in the southern states, or Nazi Germany). The proof is in the eating so to speak - that is the effects of a system on the lives of the citizens as measured by their *potential* rather than actual and subjective preferences (as stipulated by both consent and free market doctrines).
Of course, there is the question how to determine the potential without asking for the current subjective opinions, but that question can be answered empirically by finding real life examples of potential actualized in the lives of individuals and using that as models. This, of course, assumes that while different individuals have different potentials, the variation among those potentials is not that great and can be subsumed to a manageable number of types.
The bottom line here is that it is possible, and even desirable, to create a "correct" political system that is NOT based on the popular consent *prior* to its implementation. However, the issue of consent can be re-phrased in this context by stipulating that the population will eventually consent to a form of government after it experiences it and can assess its effects on the people's lives. The latter makes far more sense than asking people opinions about something they have not yet experienced - as the standard consensus approach stipulates.
Wojtek