WEBVTT 00:00:00.000 --> 00:00:03.080 Picture this. It's the afternoon of December 00:00:03.080 --> 00:00:06.919 7, 1941. You're standing in this grand office 00:00:06.919 --> 00:00:10.140 in Washington, D .C., watching what is essentially 00:00:10.140 --> 00:00:14.800 a spectacularly surreal piece of political theater. 00:00:15.000 --> 00:00:17.179 Oh, you mean in Cordell Hall's office? Exactly. 00:00:17.339 --> 00:00:19.920 The U .S. Secretary of State. So two Japanese 00:00:19.920 --> 00:00:22.760 diplomats have just walked in and they are carrying 00:00:22.760 --> 00:00:25.500 this formal 14 -part message that officially 00:00:25.500 --> 00:00:28.260 breaks off diplomatic negotiations. Right. A 00:00:28.260 --> 00:00:30.739 massive, historically tense moment. But here 00:00:30.739 --> 00:00:32.799 is the mind -bending twist to this whole scene. 00:00:33.380 --> 00:00:36.020 Secretary Hall has to act completely surprised 00:00:36.020 --> 00:00:38.539 by this, even though he has already read the 00:00:38.539 --> 00:00:40.659 fully translated message before the Japanese 00:00:40.659 --> 00:00:42.579 diplomats even finished typing it. Yeah, he was 00:00:42.579 --> 00:00:44.600 putting on like the performance of a lifetime 00:00:44.600 --> 00:00:47.100 because behind the scenes, the U .S. had already 00:00:47.100 --> 00:00:49.159 achieved something the Japanese government considered 00:00:49.159 --> 00:00:51.700 mathematically impossible. And that is exactly 00:00:51.700 --> 00:00:53.700 what we are getting into today. Welcome to this 00:00:53.700 --> 00:00:56.039 deep dive. Our mission today is to explore the 00:00:56.039 --> 00:00:58.200 allied crypt analysis project known as Magic. 00:00:58.579 --> 00:01:01.929 It's such a wild story. It really is. We're going 00:01:01.929 --> 00:01:04.730 to use our sources to unpack how the United States 00:01:04.730 --> 00:01:07.689 managed to break Japan's most secure diplomatic 00:01:07.689 --> 00:01:11.349 codes, the honestly bizarre bureaucratic turf 00:01:11.349 --> 00:01:13.689 wars that crippled the use of that intelligence, 00:01:14.129 --> 00:01:16.890 and the really controversial impacts of the secrets 00:01:16.890 --> 00:01:19.310 they uncovered. I like to think about it like 00:01:19.310 --> 00:01:21.549 having a crystal ball that shows you all of your 00:01:21.549 --> 00:01:24.750 enemy's deepest secrets, but you're only allowed 00:01:24.750 --> 00:01:27.090 to look at it every other Tuesday, and you have 00:01:27.090 --> 00:01:29.450 to read it in a locked closet. That captures 00:01:29.450 --> 00:01:31.930 the sheer frustration of the whole Magic Project 00:01:31.930 --> 00:01:35.250 perfectly. Yeah. I mean, to understand how Cordell 00:01:35.250 --> 00:01:37.349 Hull got that message early on the day of Pearl 00:01:37.349 --> 00:01:40.810 Harbor, we really need to trace this long, invisible 00:01:40.810 --> 00:01:42.790 war of mathematics. Right, because this didn't 00:01:42.790 --> 00:01:45.290 just happen overnight. Exactly. The U .S. had 00:01:45.290 --> 00:01:47.870 been chipping away at Japanese cryptography for 00:01:47.870 --> 00:01:50.290 decades at that point. The sources point back 00:01:50.290 --> 00:01:53.950 to a project in 1923 codenamed R .A .D. I saw 00:01:53.950 --> 00:01:56.510 red in the outline, but I mean, how did we even 00:01:56.510 --> 00:01:58.549 get our hands on that initial intelligence back 00:01:58.549 --> 00:02:00.810 in the 20s? Well, it actually started with physical 00:02:00.810 --> 00:02:04.310 theft. Wait, really? Yeah. A U .S. Navy officer 00:02:04.310 --> 00:02:07.209 managed to acquire a stolen copy of a Japanese 00:02:07.209 --> 00:02:09.789 naval code book that was originally used during 00:02:09.789 --> 00:02:12.539 World War I. So they photographed it. brought 00:02:12.539 --> 00:02:15.139 it back to a newly formed research desk in Washington. 00:02:15.639 --> 00:02:18.280 And they basically kept the process code in these 00:02:18.280 --> 00:02:20.860 red colored folders. Hence the name. But that 00:02:20.860 --> 00:02:23.599 seems like a massive lucky break. Just, you know, 00:02:23.780 --> 00:02:26.379 swiping the rulebook. It was huge. It gave American 00:02:26.379 --> 00:02:29.539 cryptanalysts a vital foothold. But by 1930, 00:02:29.620 --> 00:02:31.379 the Japanese government realized they needed 00:02:31.379 --> 00:02:33.960 better security. So they introduced a more complex 00:02:33.960 --> 00:02:37.719 system code named BLUE. OK, so they level up. 00:02:37.879 --> 00:02:39.979 Right. But the U .S. actually managed to break 00:02:39.979 --> 00:02:43.400 BLUE by 1932. So suddenly, American military 00:02:43.400 --> 00:02:45.680 listening stations are actively monitoring Japanese 00:02:45.680 --> 00:02:48.039 fleet commands and ship -to -ship communications. 00:02:48.139 --> 00:02:50.219 Which is a huge advantage. It is, until everything 00:02:50.219 --> 00:02:53.639 hits a brick wall in 1939. Japan's ally, Germany, 00:02:53.979 --> 00:02:55.840 sent technical assistance to upgrade Japanese 00:02:55.840 --> 00:02:58.139 cryptography. They basically provided them with 00:02:58.139 --> 00:03:00.599 modified versions of the infamous Enigma machines. 00:03:00.879 --> 00:03:03.479 Oh, wow. The Enigma machine. Yeah. And the result 00:03:03.479 --> 00:03:05.680 was this brand new system the Americans codenamed 00:03:05.680 --> 00:03:08.349 Purple Lee. Which means Purple A wasn't just 00:03:08.349 --> 00:03:10.250 another physical codebook someone could steal 00:03:10.250 --> 00:03:13.110 from a desk. Exactly. The game fundamentally 00:03:13.110 --> 00:03:17.110 changed from manual cryptography to machine -generated 00:03:17.110 --> 00:03:20.550 ciphers. With Red and BLU, you were dealing with 00:03:20.550 --> 00:03:23.090 additive or substitution codes. Like a decoder 00:03:23.090 --> 00:03:24.870 ring kind of thing. Yeah, think of it like a 00:03:24.870 --> 00:03:27.430 giant static crossword puzzle where the letter 00:03:27.430 --> 00:03:31.039 A always means attack. Humans can eventually 00:03:31.039 --> 00:03:32.979 spot the patterns and just solve the puzzle, 00:03:33.360 --> 00:03:36.580 but purpley was a machine. It began its communications 00:03:36.580 --> 00:03:39.259 with a standard line of code, but then these 00:03:39.259 --> 00:03:41.680 mechanical rotors turned the rest of the message 00:03:41.680 --> 00:03:45.180 into an unfathomable jumble. Because the rotors 00:03:45.180 --> 00:03:47.919 moved with every single keystroke, A might mean 00:03:47.919 --> 00:03:50.979 attack on Tuesday, but on Wednesday, A might 00:03:50.979 --> 00:03:53.759 mean supply ship. So code breakers tried to solve 00:03:53.759 --> 00:03:56.000 it by hand and just completely failed. Totally 00:03:56.000 --> 00:03:58.199 failed. The cipher was completely dynamic. And 00:03:58.199 --> 00:04:00.439 this is where William Friedman and Frank Rollett 00:04:00.439 --> 00:04:03.900 enter the story. They led this team of U .S. 00:04:04.039 --> 00:04:06.460 technicians who somehow, and this still blows 00:04:06.460 --> 00:04:09.580 my mind, managed to reverse engineer a machine 00:04:09.580 --> 00:04:13.620 to read purple lolly by 1939 without ever laying 00:04:13.620 --> 00:04:15.780 eyes on the original Japanese hardware. It's 00:04:15.780 --> 00:04:18.699 incredible. Imagine perfectly replicating a complex 00:04:18.699 --> 00:04:22.139 multi -tumbler lock just by listening to the 00:04:22.139 --> 00:04:24.699 clicks of someone else turning the dial all the 00:04:24.699 --> 00:04:27.939 way across the ocean. But mechanically speaking, 00:04:28.279 --> 00:04:30.759 how is that actually possible? Well, to build 00:04:30.759 --> 00:04:32.980 on your vault analogy, they didn't really need 00:04:32.980 --> 00:04:35.100 to see the physical lock if they could deduce 00:04:35.100 --> 00:04:37.699 the internal wiring. They used what cryptographers 00:04:37.699 --> 00:04:41.579 call cribs. Cribs, like predictable pieces of 00:04:41.579 --> 00:04:44.519 plain text. Exactly. The Japanese diplomats were 00:04:44.519 --> 00:04:47.110 highly formal. So their messages often began 00:04:47.110 --> 00:04:50.050 with standard, identical greetings. Oh, like 00:04:50.050 --> 00:04:52.610 to the honorable so -and -so. Right. So Friedman 00:04:52.610 --> 00:04:54.490 and Rowlett's team looked at the scrambled letters 00:04:54.490 --> 00:04:56.709 coming out of the interception, compared them 00:04:56.709 --> 00:04:58.329 to the polite greetings they knew were going 00:04:58.329 --> 00:05:00.910 into the machine on the other end, and used pure 00:05:00.910 --> 00:05:03.389 mathematical deduction to map the internal wiring. 00:05:03.509 --> 00:05:05.550 That is a genius. They essentially built their 00:05:05.550 --> 00:05:07.790 own analog of the Poole -Peele machine that could 00:05:07.790 --> 00:05:10.269 decrypt the intercepts. And this breakthrough 00:05:10.269 --> 00:05:12.560 was the birth of what they called magic. Which 00:05:12.560 --> 00:05:15.839 sounds like this incredible sweeping triumph. 00:05:16.279 --> 00:05:18.220 I mean, once you crack the unbreakable code, 00:05:18.300 --> 00:05:20.600 you'd assume the government would build a streamlined, 00:05:20.980 --> 00:05:22.720 highly efficient system to act on it. It would 00:05:22.720 --> 00:05:24.439 think so. But instead, the system they built 00:05:24.439 --> 00:05:28.100 to handle magic was astonishingly dysfunctional. 00:05:28.579 --> 00:05:31.740 You really have to factor in the fierce interservice 00:05:31.740 --> 00:05:34.240 rivalry between the U .S. Army and the U .S. 00:05:34.360 --> 00:05:37.560 Navy at the time. It was bad. Both branches were 00:05:37.560 --> 00:05:40.379 deeply paranoid about losing influence and both 00:05:40.379 --> 00:05:42.980 insisted that they alone should handle all decrypted 00:05:42.980 --> 00:05:45.920 traffic delivery to top level policymakers. It's 00:05:45.920 --> 00:05:48.620 like two kids fighting over a toy. The infighting 00:05:48.620 --> 00:05:50.779 was so bitter they reached a compromise that 00:05:50.860 --> 00:05:53.579 actively sabotaged their own intelligence gathering. 00:05:53.839 --> 00:05:56.579 The alternating days thing. They decided to literally 00:05:56.579 --> 00:05:59.160 alternate days. The army handled the decrypts 00:05:59.160 --> 00:06:01.180 one day and the Navy handled them the next. Wait, 00:06:01.560 --> 00:06:03.420 physically how does that even work? It was a 00:06:03.420 --> 00:06:06.379 mess. If a message is intercepted at 11, 5, 9 00:06:06.379 --> 00:06:09.540 p .m. on a Tuesday, who translates it? Did they 00:06:09.540 --> 00:06:11.759 literally just hand a half -translated piece 00:06:11.759 --> 00:06:14.019 of paper over at midnight like a divorced couple 00:06:14.019 --> 00:06:17.009 sharing custody? Well... was based on the data 00:06:17.009 --> 00:06:20.189 of the intercept, but the effect was just as 00:06:20.189 --> 00:06:22.730 absurd as you're imagining. Intelligence analysis 00:06:22.730 --> 00:06:25.029 requires long -term pattern recognition. Right, 00:06:25.089 --> 00:06:27.629 you have to see the big picture. Yeah, so if 00:06:27.629 --> 00:06:30.370 you are an Army officer deciding what is important 00:06:30.370 --> 00:06:33.269 to pass along on a Tuesday, you have a massive 00:06:33.269 --> 00:06:36.110 blind spot regarding the seemingly innocuous 00:06:36.110 --> 00:06:38.589 messages the Navy translated on Monday. Because 00:06:38.589 --> 00:06:41.430 you never saw them. Exactly. And the pipeline 00:06:41.430 --> 00:06:44.649 to even get a message to that desk was grueling. 00:06:44.850 --> 00:06:47.389 First came interception. The military set up 00:06:47.389 --> 00:06:50.170 these remote listening posts in places like Guam, 00:06:50.550 --> 00:06:52.870 Dutch Harbor in Alaska, Bainbridge Island in 00:06:52.870 --> 00:06:55.730 Washington state. These stations captured the 00:06:55.730 --> 00:06:58.189 raw wireless transmissions. And some of the sources 00:06:58.189 --> 00:07:00.930 mentioned they were literally mailing these intercepts 00:07:00.930 --> 00:07:03.970 to Washington, putting top -secret, time -sensitive 00:07:03.970 --> 00:07:06.250 raw data in the regular postal system. I know, 00:07:06.250 --> 00:07:08.449 it sounds crazy now, but physical mail was sometimes 00:07:08.449 --> 00:07:10.550 the only secure way to move the cipher groups 00:07:10.550 --> 00:07:12.550 without them getting intercepted again. So once 00:07:12.550 --> 00:07:15.750 they finally arrived in DC, the analog machines 00:07:15.750 --> 00:07:18.430 Friedman and Rowlett built would decipher the 00:07:18.430 --> 00:07:21.889 text. Then came translation, which was arguably 00:07:21.889 --> 00:07:24.069 the hardest bottleneck of all. Because Japanese 00:07:24.069 --> 00:07:27.779 is so context -heavy. Exactly. The Japanese language 00:07:27.779 --> 00:07:30.439 relies heavily on context and cultural nuance. 00:07:31.319 --> 00:07:33.860 The analysts didn't just need to be fluent. They 00:07:33.860 --> 00:07:36.079 needed to understand the geopolitical subtext 00:07:36.079 --> 00:07:39.079 of what the diplomats were implying. That's a 00:07:39.079 --> 00:07:41.319 huge ask. And because the Navy had invested more 00:07:41.319 --> 00:07:43.879 heavily in Japanese speaking officers, the heavy 00:07:43.879 --> 00:07:46.259 burden of translation almost entirely fell on 00:07:46.259 --> 00:07:48.879 them, regardless of whose day it was. Naturally. 00:07:49.100 --> 00:07:51.879 And then the final phase, distribution. Just 00:07:51.879 --> 00:07:54.740 sounds completely paranoid. The duty officer 00:07:54.740 --> 00:07:57.120 would pick what they personally thought was important. 00:07:57.540 --> 00:07:59.459 They would lock the decrypts in a briefcase, 00:08:00.100 --> 00:08:03.000 chain that briefcase to a relatively junior courier, 00:08:03.360 --> 00:08:05.139 and send them to make the rounds. Right, chain 00:08:05.139 --> 00:08:07.220 to their wrist. And the policymakers were shown 00:08:07.220 --> 00:08:09.439 the document briefly. They were forbidden from 00:08:09.439 --> 00:08:11.360 making copies. They couldn't take notes. And 00:08:11.360 --> 00:08:13.060 then the courier just took the briefcase right 00:08:13.060 --> 00:08:16.199 back. So policymakers were viewing these decrypts 00:08:16.199 --> 00:08:19.160 in total isolation. They couldn't cross -reference 00:08:19.160 --> 00:08:21.899 them with other intelligence or sleep on the 00:08:21.899 --> 00:08:24.000 information. Which is the worst way to analyze 00:08:24.000 --> 00:08:26.379 intelligence. And this highly fractured system 00:08:26.379 --> 00:08:29.540 directly feeds into the most famous mystery surrounding 00:08:29.540 --> 00:08:32.379 magic. If the United States was reading Japanese 00:08:32.379 --> 00:08:35.240 communications, why didn't it prevent the attack 00:08:35.240 --> 00:08:37.789 on Pearl Harbor? That is always the very first 00:08:37.789 --> 00:08:40.190 question people ask. If we broke the code in 00:08:40.190 --> 00:08:43.750 1939, why were our battleships caught completely 00:08:43.750 --> 00:08:46.809 off guard in 1941? Because you imagine you are 00:08:46.809 --> 00:08:49.309 an intelligence analyst during this time reading 00:08:49.309 --> 00:08:51.450 all these decoded messages, assuming you have 00:08:51.450 --> 00:08:54.090 the full picture. But the blind spot was enormous. 00:08:54.490 --> 00:08:56.509 And the answer actually lies in the internal 00:08:56.509 --> 00:08:58.750 politics of the Japanese government itself. OK, 00:08:58.789 --> 00:09:01.289 how so? Well, prior to Yale was a diplomatic 00:09:01.289 --> 00:09:03.509 code used exclusively by the Japanese Foreign 00:09:03.509 --> 00:09:06.740 Office. But policy and pre - Japan was largely 00:09:06.740 --> 00:09:09.940 dictated by nationalist military groups, specifically 00:09:09.940 --> 00:09:12.759 the Imperial Wave Act. Right, the military hardliners. 00:09:13.039 --> 00:09:16.159 Yeah, this was a radical ultra nationalist clique 00:09:16.159 --> 00:09:19.360 within the military that viewed civilian politicians 00:09:19.360 --> 00:09:22.440 with deep contempt. They consider their own diplomats 00:09:22.440 --> 00:09:25.659 insufficiently reliable and prone to compromising 00:09:25.659 --> 00:09:28.299 with the West. So they didn't trust them? Not 00:09:28.299 --> 00:09:30.789 at all. Therefore, the military deliberately 00:09:30.789 --> 00:09:33.169 kept the Foreign Office completely in the dark 00:09:33.169 --> 00:09:35.789 about tactical strikes like Pearl Harbor. Oh, 00:09:35.789 --> 00:09:38.409 wow. So even if the US analysts were reading 00:09:38.409 --> 00:09:40.690 every single word the Japanese Foreign Office 00:09:40.690 --> 00:09:43.230 said, the detailed operational plans for the 00:09:43.230 --> 00:09:45.330 Pearl Harbor attacks simply weren't in those 00:09:45.330 --> 00:09:48.169 messages. They literally didn't exist in the 00:09:48.169 --> 00:09:50.929 diplomatic traffic. American analysts could only 00:09:50.929 --> 00:09:53.929 draw vague inferences about deteriorating relations. 00:09:54.190 --> 00:09:56.850 But wait, if the Japanese military knew they 00:09:56.850 --> 00:09:58.649 were keeping their own foreign office out of 00:09:58.649 --> 00:10:00.330 the loop, didn't they realize that the U .S. 00:10:00.409 --> 00:10:02.090 might be listening to the diplomats and getting 00:10:02.090 --> 00:10:05.029 a totally skewed view of Japan's actual war plans? 00:10:05.370 --> 00:10:07.389 That's the thing. The historical consensus is 00:10:07.389 --> 00:10:10.009 that the Japanese military command was universally 00:10:10.009 --> 00:10:12.149 convinced that Per Peeley was mathematically 00:10:12.149 --> 00:10:15.149 unbreakable. Exactly. They didn't think the U 00:10:15.149 --> 00:10:17.409 .S. was getting a skewed view. They genuinely 00:10:17.409 --> 00:10:19.590 believe the U .S. was getting no view at all. 00:10:19.710 --> 00:10:23.110 That is wild. And, you know, while Per Peeley 00:10:23.110 --> 00:10:25.190 failed to to provide a warning for Pearl Harbor, 00:10:25.730 --> 00:10:28.549 it actually led to Magic's greatest, most unexpected 00:10:28.549 --> 00:10:31.590 triumph. Surprisingly, that triumph wasn't even 00:10:31.590 --> 00:10:33.889 in the Pacific Theater, it was in Europe. The 00:10:33.889 --> 00:10:37.309 Berlin Gold Mine. The Japanese ambassador to 00:10:37.309 --> 00:10:41.129 Germany, Baron Hiroshi Shuma, he was using his 00:10:41.129 --> 00:10:43.590 purpley machine to send daily reports back to 00:10:43.590 --> 00:10:46.289 Tokyo. And Shuma was remarkably well connected 00:10:46.289 --> 00:10:48.470 within the highest echelons of Nazi Germany. 00:10:48.649 --> 00:10:50.909 I mean, he wasn't just attending state dinners. 00:10:50.970 --> 00:10:53.889 He was sending incredibly detailed reports about 00:10:53.889 --> 00:10:56.830 his personal one -on -one strategy sessions with 00:10:56.830 --> 00:10:59.889 Adolf Hitler. Which is just insane access. He 00:10:59.889 --> 00:11:02.370 even sent dispatches detailing his guided tours 00:11:02.370 --> 00:11:04.529 of the invasion defenses in northern France, 00:11:05.090 --> 00:11:07.190 the exact beaches the Allies would later storm 00:11:07.190 --> 00:11:09.820 on D -Day. So General George Marshall considered 00:11:09.820 --> 00:11:12.100 Schumer to be his primary source of intelligence 00:11:12.100 --> 00:11:14.919 on Hitler's intentions in Europe. By looking 00:11:14.919 --> 00:11:16.899 over the shoulder of the Japanese ambassador, 00:11:17.179 --> 00:11:19.360 the Allies were effectively reading Hitler's 00:11:19.360 --> 00:11:21.899 mind. The strategic value of that just cannot 00:11:21.899 --> 00:11:25.799 be overstated. And alongside the diplomatic perpially 00:11:25.799 --> 00:11:28.820 decrypts, Allied cryptanalysts were simultaneously 00:11:28.820 --> 00:11:31.679 breaking military codes, too. Right. The Navy 00:11:31.679 --> 00:11:34.039 codes. Yeah. The U .S. Navy managed to crack 00:11:34.039 --> 00:11:37.039 into the Japanese fleet code known as JN -25. 00:11:37.320 --> 00:11:40.100 And that specific breakthrough provided the precise 00:11:40.100 --> 00:11:42.059 intelligence needed to win the battles of the 00:11:42.059 --> 00:11:44.659 Coral Sea and Midway. Which permanently crippled 00:11:44.659 --> 00:11:47.379 Japanese naval air power early in the war. Exactly. 00:11:47.659 --> 00:11:50.720 And later on, JN -25 decrypts provided the exact 00:11:50.720 --> 00:11:53.539 flight schedule of Admiral Isoroku Yamamoto, 00:11:54.000 --> 00:11:56.100 the architect of the Pearl Harbor attack, which 00:11:56.100 --> 00:11:58.179 allowed American fighter pilots to ambush his 00:11:58.179 --> 00:12:01.039 plane. But with all of this world -changing intelligence 00:12:01.039 --> 00:12:04.039 flowing in, keeping it a secret had to be practically 00:12:04.039 --> 00:12:06.779 impossible. I mean, Washington D .C. has a reputation 00:12:06.779 --> 00:12:09.179 for leaking like a sieve. Despite the locked 00:12:09.179 --> 00:12:12.039 briefcases and the chains, magic basically became 00:12:12.039 --> 00:12:14.539 the worst kept secret in town. The fact that 00:12:14.539 --> 00:12:16.720 the operation survived is just baffling luck. 00:12:16.840 --> 00:12:18.799 They survived multiple catastrophic security 00:12:18.799 --> 00:12:21.519 breaches. Let's look at the 1944 presidential 00:12:21.519 --> 00:12:24.059 election. You have the Republican candidate, 00:12:24.059 --> 00:12:27.539 Thomas Dewey, and he is planning to use Pearl 00:12:27.539 --> 00:12:30.899 Harbor as a major campaign weapon against Franklin 00:12:30.899 --> 00:12:33.159 D. Roosevelt. Right. Dewey wants to argue that 00:12:33.159 --> 00:12:35.639 the administration was incompetent and ignored 00:12:35.639 --> 00:12:38.279 warnings. So General Marshall realizes that if 00:12:38.279 --> 00:12:40.840 Dewey makes a massive public spectacle out of 00:12:40.840 --> 00:12:43.360 what the government knew before the attack, the 00:12:43.360 --> 00:12:45.240 Japanese will immediately realize their codes 00:12:45.240 --> 00:12:47.940 are compromised. It would be disastrous. So Marshall 00:12:47.940 --> 00:12:50.399 takes this extraordinary risk. He sends a highly 00:12:50.399 --> 00:12:52.929 classified personal letter directly to Dewey. 00:12:53.009 --> 00:12:55.389 Just lays it all out. Yeah, he lays out the truth. 00:12:55.590 --> 00:12:57.830 He says the U .S. has broken the codes, victories 00:12:57.830 --> 00:13:00.289 like Coral Sea and Midway were won because of 00:13:00.289 --> 00:13:02.649 this intelligence, and he practically implores 00:13:02.649 --> 00:13:05.429 Dewey to keep quiet for the sake of the war effort. 00:13:05.590 --> 00:13:08.289 And Dewey actually complies, prioritizing the 00:13:08.289 --> 00:13:10.570 war over his own campaign. But then you have 00:13:10.570 --> 00:13:12.730 the Chicago Tribune incident, which is just pure 00:13:12.730 --> 00:13:15.080 chaos. No, that was a nightmare. Just after the 00:13:15.080 --> 00:13:18.179 Battle of Midway in 1942, a reporter named Stanley 00:13:18.179 --> 00:13:20.440 Johnston effectively publishes a front -page 00:13:20.440 --> 00:13:23.559 story revealing that the U .S. had advanced knowledge 00:13:23.559 --> 00:13:26.240 of the Japanese fleet's composition and movements. 00:13:26.600 --> 00:13:28.940 Yeah, Johnston had been embedded on a transport 00:13:28.940 --> 00:13:32.259 ship and somehow managed to see an unencrypted 00:13:32.259 --> 00:13:35.120 dispatch from Admiral Nimitz that detailed the 00:13:35.120 --> 00:13:37.960 Japanese forces. And he just uses it. He used 00:13:37.960 --> 00:13:41.159 that specific data to write a series of stories 00:13:41.159 --> 00:13:43.600 for the Tribune. The level of detail in the newspaper 00:13:43.600 --> 00:13:47.700 was basically a giant glaring neon sign pointing 00:13:47.700 --> 00:13:50.320 to the fact that the JN -25 code was compromised. 00:13:50.960 --> 00:13:53.320 But somehow, and this is the crazy part, neither 00:13:53.320 --> 00:13:55.580 the Japanese government nor any of their spies 00:13:55.580 --> 00:13:58.539 noticed the Tribune coverage. They missed the 00:13:58.539 --> 00:14:00.559 follow -up stories in other papers, and they 00:14:00.559 --> 00:14:02.759 completely missed the fact that American politicians 00:14:02.759 --> 00:14:05.259 were publicly debating the leak on the floor 00:14:05.259 --> 00:14:07.259 of the United States Congress. They were utterly 00:14:07.259 --> 00:14:09.440 blind to it. They just kept clinging to their 00:14:09.440 --> 00:14:12.169 belief in the code's infallibility. Meanwhile, 00:14:12.389 --> 00:14:14.090 the circle of people who knew the secret was 00:14:14.090 --> 00:14:16.750 growing exponentially back in the US. Ranked 00:14:16.750 --> 00:14:18.889 file sailors knew the code was broken before 00:14:18.889 --> 00:14:22.309 Midway. Submarine captains, like I .J. Gallanton, 00:14:22.769 --> 00:14:24.809 openly knew they were being directed to enemy 00:14:24.809 --> 00:14:27.409 targets based on intercepted codes, which they 00:14:27.409 --> 00:14:30.590 casually just referred to as ULTRA. General Marshall 00:14:30.590 --> 00:14:32.990 investigated the security protocols at one point 00:14:32.990 --> 00:14:36.490 and discovered that over 500 people were casually 00:14:36.490 --> 00:14:38.730 reading magic intercepts at the White House. 00:14:39.070 --> 00:14:41.590 It's hilarious, honestly. Keeping a secret of 00:14:41.590 --> 00:14:45.230 this magnitude among 500 people in Washington 00:14:45.230 --> 00:14:48.669 DC might actually be the greatest magic trick 00:14:48.669 --> 00:14:51.850 of the entire war. It defies all statistical 00:14:51.850 --> 00:14:54.840 probability. You know, most Japanese cryptographers 00:14:54.840 --> 00:14:57.220 went to their graves refusing to believe Purple 00:14:57.220 --> 00:15:00.059 Lee had been broken by analytic means. They assumed 00:15:00.059 --> 00:15:02.299 there had to be a human trader passing documents. 00:15:02.519 --> 00:15:04.139 Because the math was supposed to be impossible. 00:15:04.340 --> 00:15:07.480 Right. The ultimate irony of the entire cryptographic 00:15:07.480 --> 00:15:10.000 war is that the Japanese only learned the Purple 00:15:10.000 --> 00:15:12.980 Lee cipher had been broken during post -war congressional 00:15:12.980 --> 00:15:15.139 hearings in the United States. And they were 00:15:15.139 --> 00:15:17.240 still actively using the Purple Lee machine at 00:15:17.240 --> 00:15:19.220 that time under the direction of the American 00:15:19.220 --> 00:15:21.620 Occupation Government. We were literally still 00:15:21.620 --> 00:15:23.700 reading their mail while occupying their country. 00:15:23.820 --> 00:15:26.779 It's unbelievable. But, you know, if 500 people 00:15:26.779 --> 00:15:28.799 in Washington were reading these secrets, we 00:15:28.799 --> 00:15:31.159 do have to ask how this intelligence influenced 00:15:31.159 --> 00:15:34.039 decisions at home. And this brings us to a really 00:15:34.039 --> 00:15:36.639 complex historical debate surrounding Executive 00:15:36.639 --> 00:15:40.639 Order 9066, which led to the forced relocation 00:15:40.639 --> 00:15:43.299 and internment of Japanese Americans. Now, when 00:15:43.299 --> 00:15:46.039 examining this, it is crucial we focus strictly 00:15:46.039 --> 00:15:48.519 on this historiographical debate laid out in 00:15:48.519 --> 00:15:51.169 the declassified source. rather than taking a 00:15:51.169 --> 00:15:53.389 definitive stance on motive. Absolutely. We are 00:15:53.389 --> 00:15:55.350 just reporting what the sources say here. Right. 00:15:55.669 --> 00:15:57.889 So the sources present a very active debate among 00:15:57.889 --> 00:16:01.049 historians regarding what role, if any, the magic 00:16:01.049 --> 00:16:03.769 intercepts played in the Roosevelt administration's 00:16:03.769 --> 00:16:07.070 justification for issuing Executive Order 9066, 00:16:07.490 --> 00:16:10.220 February 1942. Because when you look at the actual 00:16:10.220 --> 00:16:13.620 declassified cables from late 1941, Tokyo is 00:16:13.620 --> 00:16:15.879 blatantly ordering its consulates in places like 00:16:15.879 --> 00:16:18.840 Los Angeles and Seattle to build intelligence 00:16:18.840 --> 00:16:21.059 networks. Yes. They specifically instruct their 00:16:21.059 --> 00:16:24.120 diplomats to utilize second generations, meaning 00:16:24.120 --> 00:16:26.840 second generation Japanese Americans, alongside 00:16:26.840 --> 00:16:29.779 other groups like U .S. citizens of foreign extraction, 00:16:30.279 --> 00:16:32.919 labor union members, and anti -Semites to gain 00:16:32.919 --> 00:16:35.399 access to factories and military installations. 00:16:36.179 --> 00:16:39.009 And the declassified replies, from those consulates 00:16:39.009 --> 00:16:41.529 confirmed they were actually attempting to execute 00:16:41.529 --> 00:16:44.710 those orders. For example, the Los Angeles consulate 00:16:44.710 --> 00:16:47.570 reported establishing contacts to watch aircraft 00:16:47.570 --> 00:16:50.389 shipments in San Diego, and the Seattle consulate 00:16:50.389 --> 00:16:52.889 reported on efforts to collect intelligence on 00:16:52.889 --> 00:16:55.049 the Bremerton Naval Yard and the Boeing Airplane 00:16:55.049 --> 00:16:57.669 Factory, noting arrangements with labor union 00:16:57.669 --> 00:17:00.409 members and second -generation draftees. So the 00:17:00.409 --> 00:17:02.710 sources outline two completely opposing viewpoints 00:17:02.710 --> 00:17:05.190 on how these specific cables connect to the internment 00:17:05.190 --> 00:17:07.710 policy. On one side you have researchers and 00:17:07.710 --> 00:17:09.849 authors like David Loman and Michelle Malkin 00:17:09.849 --> 00:17:12.869 who argue that these specific magic intercepts 00:17:12.869 --> 00:17:15.470 provided the primary highly classified justification 00:17:15.470 --> 00:17:19.230 for executive order 9066. Their perspective really 00:17:19.230 --> 00:17:22.829 hinges on the fact that top officials in Roosevelt's 00:17:22.829 --> 00:17:25.410 cabinet were reading direct orders from Tokyo 00:17:25.410 --> 00:17:28.490 to recruit spies on the West Coast, and they 00:17:28.490 --> 00:17:30.410 were reading the consulate's claims of success. 00:17:31.549 --> 00:17:33.789 So according to this view, the government acted 00:17:33.789 --> 00:17:36.750 on this specific intelligence but could not publicly 00:17:36.750 --> 00:17:39.529 cite it during the war because revealing the 00:17:39.529 --> 00:17:42.829 magic intercepts would tip off the Japanese that 00:17:42.829 --> 00:17:45.690 the code was broken. which makes sense from an 00:17:45.690 --> 00:17:48.450 intelligence standpoint, but opponents of that 00:17:48.450 --> 00:17:50.930 theory point out a glaring flaw in the military 00:17:50.930 --> 00:17:53.829 chain of command regarding who actually saw the 00:17:53.829 --> 00:17:56.849 intelligence. A huge flaw. The commanding officer 00:17:56.849 --> 00:17:59.450 of the West Coast, Lieutenant General J .L. DeWitt, 00:17:59.869 --> 00:18:02.289 was the man tasked with implementing the relocation. 00:18:02.589 --> 00:18:04.890 But General DeWitt was not on the magic clearance 00:18:04.890 --> 00:18:07.730 list. He had zero access to these intercepts. 00:18:07.829 --> 00:18:10.569 None at all. Now, his superior, Secretary of 00:18:10.569 --> 00:18:12.569 War Henry Stimson, was on the clearance list. 00:18:12.950 --> 00:18:15.130 However, opponents of the intelligence justification 00:18:15.130 --> 00:18:18.029 theory point out that Stimson repeatedly requested 00:18:18.029 --> 00:18:20.450 justification for the relocation program from 00:18:20.450 --> 00:18:23.390 DeWitt. Which is the contradiction. Right. The 00:18:23.390 --> 00:18:25.609 counterargument asks a very logical question. 00:18:26.269 --> 00:18:29.470 If the magic intercepts provided the core justification 00:18:29.470 --> 00:18:32.519 for the cabinet, Why would Stimson need to ask 00:18:32.519 --> 00:18:35.200 DeWitt, who was totally blind to the intelligence, 00:18:35.779 --> 00:18:38.480 to provide further justification? Why ask the 00:18:38.480 --> 00:18:41.539 guy without the intel for the reason if you already 00:18:41.539 --> 00:18:44.180 have the intel? And further complicating the 00:18:44.180 --> 00:18:46.779 debate... When the U .S. Congress passed legislation 00:18:46.779 --> 00:18:49.980 in 1988 officially apologizing for the internment, 00:18:50.339 --> 00:18:52.180 they stated the government's actions were based 00:18:52.180 --> 00:18:55.119 on race prejudice, war hysteria, and a failure 00:18:55.119 --> 00:18:57.859 of political leadership. The congressional hearings 00:18:57.859 --> 00:19:00.299 that produced that apology did not factor the 00:19:00.299 --> 00:19:02.819 magic intercepts into their findings. So how 00:19:02.819 --> 00:19:05.440 do historians even begin to evaluate the true 00:19:05.440 --> 00:19:08.140 cause of a controversial policy when the intelligence 00:19:08.140 --> 00:19:10.539 was top secret and the key commanders on the 00:19:10.539 --> 00:19:12.220 ground weren't even permitted to see the same 00:19:12.220 --> 00:19:14.650 data as the cabinet member? historiographical 00:19:14.650 --> 00:19:17.210 nightmare, honestly. When the flow of information 00:19:17.210 --> 00:19:19.829 is artificially constrained by extreme secrecy 00:19:19.829 --> 00:19:23.009 and these rigid bureaucratic silos, determining 00:19:23.009 --> 00:19:25.690 who acted on what motive becomes incredibly difficult. 00:19:26.130 --> 00:19:28.230 Because magic was available to a very select 00:19:28.230 --> 00:19:31.369 few, the vast majority of the government, including 00:19:31.369 --> 00:19:33.410 the officers implementing policy on the ground, 00:19:34.089 --> 00:19:36.109 were operating on completely different sets of 00:19:36.109 --> 00:19:38.769 facts, assumptions, and biases. Let's try to 00:19:38.769 --> 00:19:41.380 bring all of this together. Magic wasn't just 00:19:41.380 --> 00:19:44.480 this technical triumph of code breaking. Yes, 00:19:44.640 --> 00:19:46.579 the mathematical engineering by Friedman and 00:19:46.579 --> 00:19:49.710 Rowlett was brilliant. But the deeper story here 00:19:49.710 --> 00:19:52.670 is profoundly human. Absolutely. It's a story 00:19:52.670 --> 00:19:55.930 of brilliant minds hampered by petty bureaucratic 00:19:55.930 --> 00:19:58.970 jealousy between the Army and Navy. It's a story 00:19:58.970 --> 00:20:01.549 of baffling luck and avoiding leaks when half 00:20:01.549 --> 00:20:03.769 of Washington, D .C. seemed to have access to 00:20:03.769 --> 00:20:06.589 the intercepts. And it is a story with complex 00:20:06.589 --> 00:20:09.150 domestic fallout that historians are still trying 00:20:09.150 --> 00:20:11.769 to untangle today. The legacy of magic really 00:20:11.769 --> 00:20:13.809 reminds us that technology is only as effective 00:20:13.809 --> 00:20:16.200 as the human systems built around it. I mean, 00:20:16.420 --> 00:20:18.339 the Allies achieved an intelligence miracle by 00:20:18.339 --> 00:20:21.400 reverse engineering purple E, but they nearly 00:20:21.400 --> 00:20:23.519 squandered the advantage through internal dysfunction. 00:20:23.900 --> 00:20:26.160 Which leaves you with something to really mull 00:20:26.160 --> 00:20:28.920 over. We tend to assume that gathering the right 00:20:28.920 --> 00:20:31.220 information is the hardest part of solving a 00:20:31.220 --> 00:20:34.539 crisis, but the story of magic proves that having 00:20:34.539 --> 00:20:37.700 the information means absolutely nothing if your 00:20:37.700 --> 00:20:40.440 internal systems are too fractured to understand 00:20:40.440 --> 00:20:42.400 what you're actually looking at. That's the real 00:20:42.400 --> 00:20:45.250 takeaway. Think about our modern society. We 00:20:45.250 --> 00:20:47.930 live in an age where we have more data, more 00:20:47.930 --> 00:20:49.869 signals, and more intelligence at our fingertips 00:20:49.869 --> 00:20:53.430 than at any point in human history, but are our 00:20:53.430 --> 00:20:56.190 own bureaucratic silos preventing us from connecting 00:20:56.190 --> 00:20:58.609 the dots and seeing the next Pearl Harbor before 00:20:58.609 --> 00:21:01.730 it happens? That remains the enduring cautionary 00:21:01.730 --> 00:21:04.349 lesson of the entire project. And it brings us 00:21:04.349 --> 00:21:06.569 right back to Cordell Hall, standing in his office 00:21:06.569 --> 00:21:09.829 on December 7. He possessed the ultimate crystal 00:21:09.829 --> 00:21:12.349 ball. He knew exactly what the Japanese diplomats 00:21:12.349 --> 00:21:13.950 were going to say before they even opened their 00:21:13.950 --> 00:21:16.099 mouths. But because of how that crystal ball 00:21:16.099 --> 00:21:18.619 was managed, knowing the message didn't change 00:21:18.619 --> 00:21:21.019 the outcome. Thanks for diving deep with us.