Napoleon did not recommend the study of battles, but of campaigns:
Campaigns before the twentieth century are in many ways the opposite of case studies. The latter thrusts us into a specific scenario, at a certain time and place, with a certain number of troops and resources, and with an immediate objective in mind. In set-piece battles of yore, as in modern operations, most of a commander’s mental efforts took place before things even begin: gathering an intelligence estimate of the enemy, selecting the best scheme of maneuver, matching troops to task, arranging logistics, etc. Once the action commenced, there were only a limited number of decision points where he could influence the course of events—to grossly simplify, the results came down to a series of rolls of the dice. This naturally focuses our attention on the mechanism of victory: what predetermined course of action gives the best probability of success in a scenario, accounting for a limited range of possible enemy responses.
Campaigns, on the other hand, were fundamentally more open-ended. When armies marched out of their winter cantonments for the season, they did not always know in advance which fortress to attack or where to fight a decisive action. Even when they did have a definite objective in mind, the entire challenge lay in precipitating conditions which would allow them to accomplish it; just as often, however, enemy action or the accidents of fate forced them to reconsider their intermediate steps or even the final objective itself.
This open-endedness created a very different decision-making structure. Plans had to account for a far greater degree of uncertainty, and the ability merely to remain in the field as a coherent force was at least as important as the pursuit of an objective. Sheer accident could moot one’s present course—Napoleon’s own staff was famous for issuing a steady stream of countermanding orders as circumstances evolved—and it was quite common for armies to stumble into a decisive engagement without realizing it. In short, decision-making was a far more continuous process.
This remained true even when both sides were clearly heading toward a decisive clash. Most of a commander’s focus had to remain outside the object itself: heeding his own vulnerabilities, considering the enemy’s intervening actions, ensuring that he was not detracting from his own efforts elsewhere, and getting enough men and supplies to the right place. And looming above all that was the risk of failure: what would the immediate consequences be? Was there a safe line of retreat? How would failure change his overall position? Naturally enough, the decision whether to engage in battle was just as important as the battleplan itself.
These questions shine through in narratives and memoirs from past campaigns. Many sources record war council debates over objectives, contingencies, marching routes, and supply considerations. It is by engaging with these debates and working through the decision-making process that a student can develop an intuition for the principles of war—not as a list of rules, but as a way of conducting operations in the face of uncertainty and risk. Even when the sources don’t reveal a commander’s thoughts, it can be just as instructive to try to figure out what he might have been thinking, or whether a failed action might have been somewhat justified by some non-obvious factor.
At its best military history is like a problem set with a partial answer key.
“When armies marched out of their winter cantonments for the season”
Normally armies of yore did not campaign during the winter under normal circumstances. No fodder or grazing possible for the horses.
“When armies marched out of their winter cantonments for the season” -> read it as “when winter ends and the campaign season starts”
As attributed to general Natham Bedford Forrest campaigning was/is very much “get there firstest with the mostest”