In tactics, where the enemy's measures are generally first ascertained by direct sight, and where they may be concealed by every wood, every fold of undulating ground, we must naturally always be alive, more or less, to the possibility of unforeseen events, in order to strengthen, subsequently, those points which appear too weak, and, in fact, to modify generally the disposition of our troops, so as to make it correspond better to that of the enemy.
Such cases must also happen in Strategy, because the strategic act is directly linked to the tactical.

In Strategy also many a measure is first adopted in consequence of what is actually seen, or in consequence of uncertain reports arriving from day to day, or even from hour to hour, and lastly, from the actual results of the combats it is, therefore, an essential condition of strategic command that, according to the degree of uncertainty, forces must be kept in reserve against future contingencies.
In the defensive generally, but particularly in the defence of certain obstacles of ground, like rivers, hills, &c., such contingencies, as is well known, happen constantly.
But this uncertainty diminishes in proportion as the strategic activity has less of the tactical character, and ceases almost altogether in those regions where it borders on politics.
The direction in which the enemy leads his columns to the combat can be perceived by actual sight only; where he intends to pass a river is learnt from a few preparations which are made shortly before; the line by which he proposes to invade our country is usually announced by all the newspapers before a pistol shot has been fired.

The greater the nature of the measure the less it will take the enemy by surprise.

Time and space are so considerable, the circumstances out of which the action proceeds so public and little susceptible of alteration, that the coming event is either made known in good time, or can be discovered with reasonable certainty.
On the other hand the use of a reserve in this province of Strategy, even if one were available, will always be less efficacious the more the measure has a tendency towards being one of a general nature.
We have seen that the decision of a partial combat is nothing in itself, but that all partial combats only find their complete solution in the decision of the total combat.
But even this decision of the total combat has only a relative meaning of many different gradations, according as the force over which the victory has been gained forms a more or less great and important part of the whole.

The lost battle of a corps may be repaired by the victory of the Army.