These circumstances are country and ground, the former including the territory and inhabitants of the whole theatre of war; next the time of the day, and the time of the year as well; lastly, the weather, particularly any unusual state of the same, severe frost, &c. 36. THESE FORM NEW MEANS. By bringing these things into combination with the results of a combat, strategy gives this result—and therefore the combat—a special signification, places before it a particular object. But when this object is not that which leads directly to peace, therefore a subordinate one, it is only to be looked upon as a means; and therefore in strategy we may look upon the results of combats or victories, in all their different significations, as means. The conquest of a position is such a result of a combat applied to ground. But not only are the different combats with special objects to be considered as means, but also every higher aim which we may have in view in the combination of battles directed on a common object is to be regarded as a means. A winter campaign is a combination of this kind applied to the season. There remain, therefore, as objects, only those things which may be supposed as leading _directly_ to peace, Theory investigates all these ends and means according to the nature of their effects and their mutual relations. 37. STRATEGY DEDUCES ONLY FROM EXPERIENCE THE ENDS AND MEANS TO BE EXAMINED. The first question is, How does strategy arrive at a complete list of these things? If there is to be a philosophical inquiry leading to an absolute result, it would become entangled in all those difficulties which the logical necessity of the conduct of War and its theory exclude. It therefore turns to experience, and directs its attention on those combinations which military history can furnish. In this manner, no doubt, nothing more than a limited theory can be obtained, which only suits circumstances such as are presented in history. But this incompleteness is unavoidable, because in any case theory must either have deduced from, or have compared with, history what it advances with respect to things. Besides, this incompleteness in every case is more theoretical than real. One great advantage of this method is that theory cannot lose itself in abstruse disquisitions, subtleties, and chimeras, but must always remain practical. 38. HOW FAR THE ANALYSIS OF THE MEANS SHOULD BE CARRIED.