The connection of fortification and strategy may be considered under two distinct heads:  1st, the choice of sites for constructing fortresses for defence; 2d, their influence in offensive operations, and the determination of the question whether they can be passed with safety, or whether the attacking force will be under the necessity of besieging them.

The centre and extremities of a base of operations should always be secured either by natural or artificial obstacles.  This base is generally chosen so that fortifications will be necessary for strengthening only a part of the line.  But if a frontier, like the side of France towards Belgium, be destitute of natural obstacles, the artificial means of defence must be proportionally increased.  Great care should be taken that permanent fortifications be made only on such places as may favor military operations.  If otherwise, the troops detached from the active army for garrisoning them, will only tend to weaken this force without any corresponding advantages.  In this way, fortifications may become actually injurious to defence.  A number of the European fortresses which were built before the subject of strategy was properly understood, are now regarded as utterly useless, from their ill-advised positions.

Whether a fortress may be safely passed with merely blockading or observing it, depends very much upon the nature of the war, and the numbers and position of the defensive army.  The allies, in 1814, invading France with a million of soldiers, assisted by the political diversion of factions and Bourbonists within the kingdom, and treason in the frontier fortresses, and even in the ranks of Napoleon’s army, could conduct their military operations on a very different plan from that which would be adopted by either Austria, Prussia, Russia, England, Spain, Portugal, Holland, Italy, and the German powers, if singly waging war with the French.  Napoleon sometimes detached a corps to observe a fortress which threatened his line of operations or of manoeuvre; at others, he delayed his advance till the place could be reduced.

“An army,” says Jomini, “may sometimes penetrate between places on an open frontier, to attack the enemy’s forces in the field, taking care at the same time to observe these places; but no invading army can cross a great river, like the Danube, the Rhine, or the Elbe, without reducing at least one of the fortresses on that river, so as to secure a line of retreat; but being in possession of such a place, it can continue the offensive, while its materiel de siege successively reduces the other places.”

In case the main army is obliged to remain and cover the besieging corps, it should take some central position, where it can command all the avenues of approach, and fall with vigor on the enemy, should he attempt to raise the siege.  Napoleon’s operations before Mantua, in 1796, offer the finest model for imitation.