Tuesday, September 6, 2016

More Failures of Military Intelligence

The difficulty with using even the best intelligence is underscored by this story from the Arab-Israeli wars. In 1970 Ashraf Marwan, the son-in-law of Egypt’s President Gamal Abdel Nasser and one of his closest confidants, volunteered his services to Mossad as a spy. He was code-named Angel and was for years Israel's best agent:
A new book by veteran Israeli intelligence expert Uri Bar-Joseph tells the story of the Angel, Marwan’s code name, in detail. This is a tale of espionage at the highest level. The Angel provided Israel with Egypt’s entire order of battle for its armed forces and Egypt’s war plans for attacking Israel across the Suez Canal. He provided details of Sadat’s meetings with the Soviet leaders and up-to-the-minute reports on Soviet arms deliveries to Cairo.

But the Israeli military intelligence experts in the Directorate of Military Intelligence, which was solely responsible for producing the national intelligence estimates on whether Egypt would go to war, were convinced Sadat would not take the risk. The DMI had a concept of war planning. In the concept, Egypt could not beat Israel because of Israel’s overwhelming air superiority, which Egypt’s leaders knew made war suicidal. Thus Sadat wouldn’t attack.
So despite clear warnings from their top agent the Israelis were surprised by Egypt's 1973 attack across the Suez Canal, and it took them hard fighting and good luck to avoid losing a whole infantry division to Egypt's pincer attack.

The story doesn't get any better after the war. Eli Zeira, the chief of Israel's Directorate of Military Intelligence, faced hard questioning about why he had not taken Marwan's warning seriously. He responded by launching a covert campaign to smear Marwan as an untrustworthy double agent, eventually leaking so many details about his spy that Marwan's identity was exposed by the press, leading to his murder in London in 2007. Now that's a great way to encourage future agents to risk their lives for you.

1 comment:

G. Verloren said...

The problem with intelligence is that's only information, and it still needs to be interpreted and made sense of.

Imagine you find a complete copy of an ancient text which is very similar to, but in certain ways critically different from, an already known text. Suppose you show it to an appropriate authority and argue this rewrites our understanding of history, but they believe these difference to be insignificant - perhaps they choose to believe it is an altered copy made in imitation of the text they already know. They may even believe this despite exacting and direct evidence to the contrary, such as where the text was found, the structure of the linguistics, the materials it was transcribed on, et cetera.

Bias creeps into every field of human envdeavor. It's not enough to have good information - you also need to have authorities who are able to overcome their own extant biases to accept such information even when it goes contrary to all their knowledge, experiences, and expectations. Even geniuses can be paralyzed by an unwillingness to consider positions that go against their current expertise - e.g., Einstein famously rejected quantum mechanics, despite fairly strong evidence for it.