Fortitude South: The Army That Did Not Exist

Operation Fortitude South did not have to persuade the German high command of anything. It had to prevent them from changing their minds.

German planners already believed the invasion would come at the Pas de Calais. It was the obvious choice: the shortest crossing, the closest to the Ruhr, the best road network beyond, and the location of the V-weapon sites. Rommel and von Rundstedt both expected it. The deception's task was therefore not construction but maintenance, which is a far cheaper problem.

The apparatus was the First United States Army Group, FUSAG, notionally assembled in south-east England opposite Calais. It consisted of radio traffic simulating the movement and administration of an army group, inflatable tanks and dummy landing craft positioned where aerial reconnaissance would find them, and — the decisive element — reports from the double agents of the Double-Cross system, above all GARBO.

It was commanded by George Patton, who was at that moment without a command following the slapping incident in Sicily. This was the shrewdest element of the plan. German intelligence regarded Patton as the best Allied field commander, and therefore reasoned that he would lead the main effort. The inference was made by the enemy, from a true fact, and required no persuasion at all.

The mechanism is confirmation bias, in its precise sense. Wason's rule-discovery experiments established that people test hypotheses by seeking confirming instances rather than disconfirming ones, and an intelligence service that has adopted a hypothesis will collect evidence for it faster than against it. Fortitude did not have to be convincing. It had to be available.

The part of the operation that is most instructive came after D-Day.

The Normandy landings on 6 June were, on any reading, an invasion. A rational assessment would have reallocated reserves immediately. Instead the deception continued: GARBO's 9 June message, further FUSAG traffic, and continued signals from the notional army group. The Fifteenth Army — nineteen divisions — remained at Calais for weeks, waiting for a second landing that did not exist.

That persistence is belief perseverance. Once a hypothesis has organized a body of evidence, disconfirmation does not simply cancel it; the accumulated interpretation survives, and the new event is fitted into the existing frame rather than replacing it. The Normandy landings were absorbed as the diversion the theory had already predicted.

The relevant defensive question, and it applies well outside intelligence, is not what does the evidence support. It is: what evidence would I expect to see if my current belief were wrong, and have I looked for it.