[T]here is also a kind of logical argument which can be used to show the inconsistency of any of the particular forms of the theory of sovereignty; more precisely, the logical argument can be given different but analogous forms to combat the theory that the wisest should rule, or else the theories that the best, or the law, or the majority, etc., should rule. One particular form of this logical argument is directed against a too naive version of liberalism, of democracy, and of the principle that the majority should rule; and it is somewhat similar to the well-known ‘paradox of freedom’ which has been used first, and with success, by Plato. In his criticism of democracy, and in his story of the rise of the tyrant, Plato raises implicitly the following question: What if it is the will of the people that they should not rule, but a tyrant instead? The free man, Plato suggests, may exercise his absolute freedom, first by defying the laws and ultimately by defying freedom itself and by clamouring for a tyrant. This is not just a far-fetched possibility; it has happened a number of times; and every time it has happened, it has put in a hopeless intellectual position all those democrats who adopt, as the ultimate basis of their political creed, the principle of the majority rule or a similar form of the principle of sovereignty. On the one hand, the principle they have adopted demands from them that they should oppose any but the majority rule, and therefore the new tyranny; on the other hand, the same principle demands from them that they should accept any decision reached by the majority, and thus the rule of the new tyrant. The inconsistency of their theory must, of course, paralyse their actions. Those of us democrats who demand the institutional control of the rulers by the ruled, and especially the right of dismissing the government by a majority vote, must therefore base these demands upon better grounds than a self-contradictory theory of sovereignty. [ch. 7, 134-5]
Institutions for the selection of the outstanding can hardly be devised. Institutional selection may work quite well for such purposes as Plato had in mind, namely for arresting change. But it will never work well if we demand more than that, for it will always tend to eliminate initiative and originality, and, more generally, qualities which are unusual and unexpected. This is not a criticism of political institutionalism. It only re-affirms what has been said before, that we should always prepare for the worst leaders, although we should try, of course, to get the best. But it is a criticism of the tendency to burden institutions, especially educational institutions, with the impossible task of selecting the best. This should never be made their task. This tendency transforms our educational system into a race-course, and turns a course of studies into a hurdle-race. Instead of encouraging the student to devote himself to his studies for the sake of studying, instead of encouraging in him a real love for his subject and for inquiry, he is encouraged to study for the sake of his personal career; he is led to acquire only such knowledge as is serviceable in getting him over the hurdles which he must clear for the sake of his advancement. In other words, even in the field of science, our methods of selection are based upon an appeal to personal ambition of a somewhat crude form. (It is a natural reaction to this appeal if the eager student is looked upon with suspicion by his colleagues.) The impossible demand for an institutional selection of intellectual leaders endangers the very life not only of science, but of intelligence.
It has been said, only too truly, that Plato was the inventor of both our secondary schools and our universities. I do not know a better argument for an optimistic view of mankind, no better proof of their indestructible love for truth and decency, of their originality and stubbornness and health, than the fact that this devastating system of education has not utterly ruined them. [ch. 7, 147-8]