I read Persico's Roosevelt's Centurions two years ago, during that deluge of books about FDR's military commanders my dad dropped on me. This book is all about the various aspects of espionage that went on under FDR's control before and during World War 2. The struggles with codebreaking, the attempts to capture foreign spies and saboteurs, the deliberate decision not to act on certain pieces of intelligence, the interdepartmental struggles, the high and low points of the OSS' early years.
All in all, FDR comes off as someone who sounds like he'd be horrible to work for. He very much keeps his own counsel, and is excellent at telling people what they want to hear, even if he has no intention of supporting them. He often leaves it deliberately vague where one intelligence agency's authority ends and another begins, leading to a lot of fighting, attempts to poach talent, and undermine rivals, even though ostensibly all these people are on the same side and supposed to be working for the benefit of the nation. And Roosevelt seems to relish this shit, the idea he's the only who sees the whole picture.
When I mentioned choosing to not act on certain information, I don't mean foreknowledge of the attack on Pearl Harbor (I was referring to some atrocities committed by the Soviets he chose to ignore). Persico does address that idea, but concludes there's no evidence FDR had such foreknowledge, and that overall there was little intelligence available to the U.S. that pointed at an attack there, and it was buried in amongst a lot of other intelligence pointing at Japanese offensives in other directions. He also argues that FDR wanted to fight Germany, not Japan, and thus wouldn't have been spoiling for a fight with the latter, since he couldn't count on Hitler to ignore the stated terms of his alliance with Japan to declare war on the U.S. at what was an inopportune time for Germany.
I'm not sure how that squares with Kemp Tolley's Cruise of the Lanikai, where it certainly seemed he was being sent on a largely pointless mission to spy on the Japanese so as to draw fire, which could be used as a pretext for a war (only for it all to be rendered pointless when the Japanese attacked Pearl Harbor first).
Persico only devotes a few pages to FDR's decision to intern thousands of American citizens on the basis of their Japanese ancestry. He attributes it to FDR's considerable fear of fifth columnists working against the U.S. from within. Which could be a valid concern, if he were simply arresting people actually caught in the act of committing sabotage or whatever. But to lock up an entire group of people, essentially because of some loudmouth, hysterical racists, is just stupid. Persico, while acknowledging that it's an extremely dark mark on FDR's record, does try to excuse it a bit by arguing FDR can't get too far ahead of public opinion. Which is, to be frank, bullshit. Public opinion was overwhelmingly against entering World War 2 prior to Pearl Harbor. Which didn't stop FDR from doing everything he could to aid Britain and to try and have U.S. Naval vessels essentially pick fights with U-boats, because he was convinced he was right. If he really believed the public opinion about American citizens of Japanese heritage was wrong, he could have turned on the charm and sold the public on his way of seeing things. Or just ignored them and done what he thought was right anyway, which he did plenty of times before and after. He didn't, for stupid reasons, and that's got to be held against him.
If you have an interest in the espionage side of World War 2, this could prove interesting to you. It also deals a bit with British, Soviet, and German intelligence, mostly in terms of their efforts compared to, or against the U.S. It makes for quite a few interesting stories.
'Searching for a way his industry could contribute to the war effort, Starr helped the OSS put together a small Insurance Investigation Unit, rarely numbering more than half a dozen members. The unit based its work on one of the fundamentals of the insurance business, that insurers require a detailed description of the properties they insure - their size, location, a complete physical profile. This condition included policies written on the plants of German and Japanese arms makers, the details of which would provide priceless intelligence for Allied bombers.'