It follows from them that besides a theoretical acquaintance with the subject, natural talent must also have a great influence on the value of critical examinations, for it rests chiefly with the latter to throw the requisite light on the interrelations of things, and to distinguish from amongst the endless connections of events those which are really essential.
But talent is also called into requisition in another way.

Critical examination is not merely the appreciation of those means which have been actually employed, but also of all possible means, which therefore must be suggested in the first place--that is, must be discovered; and the use of any particular means is not fairly open to censure until a better is pointed out.

Now, however small the number of possible combinations may be in most cases, still it must be admitted that to point out those which have not been used is not a mere analysis of actual things, but a spontaneous creation which cannot be prescribed, and depends on the fertility of genius.
We are far from seeing a field for great genius in a case which admits only of the application of a few simple combinations, and we think it exceedingly ridiculous to hold up, as is often done, the turning of a position as an invention showing the highest genius; still nevertheless this creative self-activity on the part of the critic is necessary, and it is one of the points which essentially determine the value of critical examination.
When Buonaparte on 30th July, 1796,( *) determined to raise the siege of Mantua, in order to march with his whole force against the enemy, advancing in separate columns to the relief of the place, and to beat them in detail, this appeared the surest way to the attainment of brilliant victories.

These victories actually followed, and were afterwards again repeated on a still more brilliant scale on the attempt to relieve the fortress being again renewed.

We hear only one opinion on these achievements, that of unmixed admiration.
(*) Compare Hinterlassene Werke, 2nd edition, vol.iv.