Command
Military Leaders: Decisions Under Uncertainty, Not Rankings
Command is a sequence of decisions made without enough information. Studying it means reconstructing what was actually knowable at the time, which is harder and more useful than ranking reputations.
Writing about military leadership divides fairly cleanly into two kinds. One ranks commanders and argues about the ordering. The other examines particular decisions and asks what information they rested on. The first is popular and tells you very little; the second is harder and is what this section attempts.
The discipline it requires is resisting hindsight. A decision that looks obvious now is usually one whose alternatives have been forgotten, and the interesting question is almost never was this right but was this reasonable given what was in front of them. An account in which the losing commander is simply foolish has skipped the work.
What the record can and cannot support
Orders, dispatches, unit diaries and inquiries record what was decided and often when. They record reasoning much less reliably, and memoirs written afterwards are shaped by knowing the outcome. Where a commander's reasoning is genuinely documented (a contemporaneous interview, a written appreciation, testimony to an inquiry) that is worth more than any later reconstruction, and those are the sources these pages lean on.
Profiles
5 commanders- 01 Washington's Strategy: Preserving the Army as the Objective He lost more engagements than he won and achieved the political objective anyway, because he saw that the army's survival was the war's centre of gravity.
- 02 Grant's Method: Continuous Pressure Instead of Battles He was called a butcher by contemporaries and is now studied for the opposite reason: he understood that the war would be won by not stopping.
- 03 Pershing and the AEF: Building a Modern Army From Almost Nothing The hard part was not commanding in France. It was raising, training, shipping and supplying millions of men from an army of a few tens of thousands.
- 04 Nimitz in the Pacific: Commanding a Campaign Across an Ocean He took over a fleet that had just been badly damaged and ran a campaign where distance, logistics and intelligence mattered more than any engagement.
- 05 Eisenhower as Coalition Commander: The Job Nobody Trained For He never commanded a division in combat. What he did was hold together an alliance of armies that had every reason to fragment, which was the harder problem.
Command decisions on this site
- Petraeus and the 2007 surge A commander explaining a strategy publicly while executing it, before the outcome was known.
- The Sixth Fleet off Beirut Command decisions in a mission defined as presence rather than as an objective.
- Pointe du Hoc Junior leaders acting without orders when the plan met the ground.
The problem of the single decisive figure
There is a structural bias in how command is written about. A campaign involves tens of thousands of people and is remembered through one or two of them, because a name is easier to carry than an organisation. That compression is not merely unfair to the unnamed; it produces bad analysis. It encourages the belief that outcomes turn on personal quality, when the recurring pattern is that commanders inherit a situation largely shaped before they arrived and are judged on how much of it they could still move.
The corrective is to ask what the commander actually controlled. Some things are genuinely theirs: the allocation of reserves, the choice of where to accept risk, the tempo they demand, whether they trust subordinates enough to let them act on incomplete orders. Other things are not: the force they were given, the terrain, the political deadline, the intelligence picture, the state of the supply line when they took over. An assessment that does not separate the two is describing circumstances and calling them character.
Why removal is a poor measure of failure
Commanders are relieved for many reasons and only sometimes for incompetence. A relief can settle a political argument, signal a change of approach to an ally, or simply mark that the previous approach has been abandoned and someone has to carry that. Conversely, commanders survive catastrophes when there is nobody available to replace them or when replacing them would concede something. Reading the personnel record as a performance record therefore misleads in both directions, and the dates of an appointment are often more informative than the reason given for its end.
The pages in this section are written with that in mind. Where a commander is credited with a decision, the aim is to say what alternatives were live at the time and who else had a hand in it. Where the record does not support a clean attribution, the page says the attribution is unclear and not picking the version that reads best.
A last practical note on reading this section. Where a commander appears in more than one page, the accounts are deliberately not reconciled into a single verdict: the same person can have made a defensible decision in one campaign and an indefensible one in another, and averaging the two into a reputation discards the information. The pages are organised by decision rather than by biography for that reason.
Questions and answers
Why are there no ranked lists of great commanders here?
Because they measure reputation as opposed to anything about command. A general who won with overwhelming material advantage and one who held a hopeless position competently are not comparable on a single axis, and ranking them tells you about the ranker. These pages examine specific decisions instead.
What makes a command decision worth studying?
That it was made under real uncertainty and can be reconstructed: what the commander knew, what they believed, what they chose and what followed. A decision that looks obvious in hindsight is usually one where the alternatives have been forgotten, and recovering those alternatives is most of the work.
Does this section cover living officers?
Where the public record supports it and the subject is a documented professional decision rather than a personal matter, yes. Contested claims are described as contested, and nothing is asserted about a living person that the record does not carry.