This idea, which is held with some variation by Sumner, Gobineau, Faguet, and Conway, is, I believe, both unhistorical and unpsychological, because it is but a half-truth. This substratum of the population does at the moment of revolution become a dangerous mob. Such people are unadjusted to any social order, and the least deviation from the routine of daily life throws them off their balance. The relaxation of authority at the moment when one group is supplanting another in position of social control, is to these people like the two or three days of interregnum between the pontificates of Julius and Leo, described by Cellini. Those who need some one to govern them, and they are many, find their opportunity in the general disturbance. They suddenly react to the revolutionary propaganda which up to this minute they have not heeded, they are controlled by revolutionary crowd-ideas in a somnambulistic manner, and like automatons carry these ideas precipitately to their deadly conclusion. But this mob is not the really revolutionary crowd and in the end it is always put back in its place by the newly dominant crowd. The really revolutionary crowd consists of the group who are near enough the dominant crowd to be able to envy its "airs" with some show of justification, and are strong enough to dare try issue with it for supreme position. Madame Rolland, it will be remembered, justified her opposition to aristocrats on the principle of equality and fraternity, but she could never forget her resentment at being made, in the home of a member of this aristocracy, to eat with the servants. What Le Bon and others seem to ignore is that the ruling class may be just as truly a crowd as the insurrectionary mob, and that the violent behavior of revolutionary crowds is simply the logic of crowd-thinking carried to its swift practical conclusion.