Closing the book on Pearl Harbor

Budiansky, Stephen

ABSTRACT: Recent attempts to resuscitate well-worn conspiracy theories concerning the Pearl Harbor attack are based on selective reading of documentary evidence and ignore conclusive, recently declassified materials which show that JN-25 and other Japanese naval codes were not being read by U. S. Navy codebreakers prior to the Japanese attack.

KEYWORDS: JN-25, Pearl Harbor, Crane Files, OP-20-G, Japanese naval codes.

The Japanese attack on Pearl Harbor, it is a safe bet, is the most studied event in American history. Even before the war had ended it had been the subject of eight boards of inquiry and investigation which sought to discover how the United States had been so unprepared and how a chain of almost unbelievable gaffes had left the U. S. Pacific Fleet sitting like a row of ducks at anchor on the bright morning of December 7, 1941. After the war, Republicans in Congress, convinced that a man who could defeat their party four times running was capable of anything, helped to launch a massive probe of Pearl Harbor with the all but stated aim of proving that Franklin Roosevelt or other high political and military officials of the government knew of the planned Japanese attack in advance and did nothing to avert it, so desperate was FDR's circle to get America into the war. To the fury of America's intelligence community, the Congressional probe and the other official investigations (whose complete transcripts were published in a massive, 39-volume series in 1946) thoroughly compromised the fact that the United States had broken the major Japanese diplomatic cipher machine, which the United states called PURPLE, well before the war. Such codebreaking successes were with much justification considered to rank among the nation's most closely guarded secrets, even after the defeat of Nazi Germany and Japan. Bitter experience had shown that revelations of success in reading another nation's codes often brought about swift improvements in cryptologic security. This was not just a theoretical concern in 1946; at that time the United States and Britain were beginning to break the Soviet diplomatic and KGB codes (the projects known as VENONA in the United States and ISCOT in Britain), and these were revealing the existence of a sizable espionage operation within the United States as well as Soviet control and manipulation of Communist groups in post-war Europe. Any leaks about Allied codebreaking prowess was thus an extremely touchy matter: Both Britain and America had kept extremely tight wraps on their codebreaking establishments even after the war's end. In that pre-whistle-blower age when faith in government still ran strong, employees at the codebreaking establishments signed a pledge that they would never at any time betray any information about their work, and they didn't.

Aside from the fear that disclosures would make their work more difficult in the future, the codebreakers had another justifiable fear about the investigations: it was one thing to acknowledge that we had been spying on the Germans and Japanese, quite another that we had, even before our entry into the war, been spying on the French, the Russians, and the Mexicans and other Latin American nations. The fact that the American codebreakers had engaged in a full sharing of ideas, methods, intercepted messages, and even some decrypts with Britain a full year before America's entry into the war was also a source of great potential embarrassment. This cooperation was a definite part of Churchill's efforts to get the United States and Britain "somewhat mixed up together," in his apt phrase, and Roosevelt heartily approved the suggestion for these and other close military-to-military contacts beginning in the summer of 1940 as Britain fought alone against Hitler.1

The decrypted Japanese diplomatic messages, which went under the U. S. codename MAGIC, were minutely scrutinized during the Congressional investigations, and have been ever more minutely scrutinized in the years since, for any hint of warnings ignored, and it is well established that on the night of December 6, FDR was handed a signal which made it abundantly clear that Japan was preparing to break off diplomatic relations, which could mean nothing but war. But, to the eternal disappointment of the conspiracy theorists, the messages contain not a hint of where Japan might strike. Conventional wisdom suggested the Philippines, and Secretary of the Navy Frank Knox was hardly alone in being flabbergasted when the news came in from Pearl Harbor: "My God, this can't be true, this must mean the Philippines!" he exclaimed. All that the MAGIC decrypts prove is the decidedly nondastardly fact that America had no choice but to wait for Japan to strike the first blow; it was inconceivable that America might launch military action preemptively, even with a clear warning that Japan was about to break off diplomatic relations.

This, of course, has not deterred a number of "revisionist" historians who have been determined at all costs to prove that FDR had to have known in advance that Pearl Harbor was the target and, knowing it, deliberately hid the information. But these narratives, however exciting, have always been rather unsatisfactory, for they depend on all sorts of convoluted tales about Japanese spies, and jargon messages slipped into Radio Tokyo broadcasts, and somebodysaid-that-somebody-said testimony from dubious sources.

One thread of continuing suspicion and excitement for the conspiracy theorists in search of something stronger, however, has always been the Japanese naval codes. Because of continuing security classification, technical details about the breaking of most Axis codes remained under wraps for a half century, with the first major, wholesale release of documents occurring only in 1996, when the National Security Agency transferred to the National Archives some 5,000 files from Arlington Hall's records. Even now some World War II documents are classified out of exaggerated fears that any discussion of codebreaking methods and techniques, even methods from the pre-computer age, would compromise national security.

The very existence of such a seemingly incomprehensible veil of secrecy, of course, arouses suspicion; and for some time now a number of revisionist authors have been raising the suggestion that what is being hidden is the fact that the United States was routinely reading not only the diplomatic messages but also the operational naval messages of the Japanese fleet before Pearl Harbor. Such operational traffic messages, it is presumed, could not fail to have delivered the goods; unlike the PURPLE traffic, which dealt only with high level diplomatic negotiations, these naval signals would contain orders to the fleet at all levels, and would surely have betrayed to even the dimmest intelligence analyst the object of Japan's military intentions.

Robert Stinnett now claims to have unearthed evidence that that was indeed the case. His book Day of Deceit (Free Press, $26.00) received a huge amount of prepublication hype, with plugs from Gore Vidal on the back cover and in the Nation,2 among other things. Stinnett recycles all of the arguments familiar to readers of the Pearl Harbor revisionists, but adds what the publisher terms an "explosive" revelation. Examining "an extraordinary number of documents whose release I have been able to engineer through Freedom of Information Act requests," the author writes, he has determined that "by the closing the months of 1941, America was intercepting and breaking-within a matter of hours-most every code that Japan could produce. . . The truth is clear: FDR knew."

Stinnett's assertions about the Japanese naval codes are the crux of his thesis, and his only real claim to advancing the conspiracy narrative beyond its familiar limits, and so it is worth a close examination of the evidence that the United States had indeed penetrated the naval codes long before the officially acknowledged date of spring 1942.

The most immediate difficulty with Stinnett's contention is that it has been directly and repeatedly contradicted by the testimony of the U. S. naval officers whose responsibility it was to break the Japanese codes. Stinnett brushes this problem aside easily enough: they were "lying" or "part of the cover up"-a "sixty year cover up [that] has hidden American and Allied success in obtaining the solutions to the naval codes before Pearl Harbor." Well, you can't have a conspiracy without conspirators, so perhaps we should grant him this point for the sake of argument.

Stinnett's most prized piece of evidence, however, is a letter from Admiral Royal Ingersoll, assistant chief of naval operations, dated October 4, 1940, in which he states that the main Japanese naval code system, the "Operations Code," was well on its way to being solved; some messages were being decoded in a matter of hours and a full solution was expected by April 1941. Stinnett states that another Navy document reveals that in the fall of 1941 "a revised solution" to the Operations Code was sent to the Navy's codebreaking unit in Hawaii, "an astonishing disclosure, for it directly contradicts postwar claims that the code was not readable until 1942 or early 1943." Stinnett admits his direct knowledge of the matter ends there; "documents concerning the (Operations Code remain classified," he states, but that in itself, he argues, is evidence of a massive cover-up: "This extraordinary secrecy, which still remains in effect in 1999, is intended to distance the American government and particularly FDR from foreknowledge of Japanese attack plans."

Stinnett has spent two decades on his project, and he has studied many original archival documents, particularly at the National Archives at College Park. He has however ignored-or, more likely, is ignorant of-key documents that not only fail to support his theories but that devastatingly undermine them. The two most important of these documents are a complete internal history of the solution of the Operations Code, formerly classified TOP SECRET ULTRA, and a series of month-by-month progress reports filed by OP-20-G throughout 1940 and 1941 which tally in precise detail how many code groups had been recovered in each of the code systems under study.3 These two documents, which are also fully corroborated by other contemporaneous reports and memoranda, show unmistakably that not a single message sent throughout the year 1941 in the Operations Code was broken and read by the United States before December 7. I will discuss the evidence contained in these documents in more detail below.

Stinnett is misled in part by a very basic misunderstanding of the nature of the Japanese codes and the process of cryptanalysis. The Japanese naval codes (and also most of the high-level Japanese army codes as well, which would be broken beginning in 1943 by Arlington Hall) were all of a type known as enciphered codes. Words, numbers, names, phrases each were assigned a unique numerical value, in the case of the Operations Code a five-digit number. A message was prepared by first translating the plaintext into a string of corresponding five-digit code groups. To each of these code groups was then added a random five-digit number. These were drawn in sequence from a book of "additives." The resulting "enciphered groups" were what were then actually transmitted. The additive or "key" book, in the case of the Operations Code, contained from 30,000 to (in later years of the war) 100,000 five-digit numbers. Both sender and recipient possessed a copy of the key book, and whenever a message was sent the sender would include within the message an "indicator," a coded number that informed his recipient the page number and line number of the additive book from which he had begun drawing his additives. This meant that the enciphered group for, say "Manila" might be 04729 in one message and 77204 in another, depending on the starting point in the key book that had been chosen for that particular message.

Breaking such a code involves really three distinct phases. The first is a purely theoretical or research phase. Large amounts of enciphered traffic are intercepted, accumulated, and searched for any clues as to what type of code it is and its method of operation. One major goal in attacking an enciphered code is to discover the system used for encoding and hiding the indicator. For once that is found, it begins to be possible to build up what cryptanalysts call a "depth" : finding messages that happened to have been enciphered using overlapping sequences of the key book. Only then can the second phase begin; this is the huge, and incredibly labor intensive task, of teasing apart the additives from the underlying code groups. This is a process that requires poring over literally tens of thousands or even hundreds of thousands of messages in search of overlaps, and then applying a series of arithmetic searches for common numerical differences from message to message. It is a mind-numbing, tedious task that can take months or years. The ultimate goal is to reconstruct the key book and to identify and figure out the linguistic meanings of the original code groups. The Japanese codes had code books which generally contained some 30,000-50,000 separate code groups.

Only when a sufficient fraction of the code book and key book has been recovered in this manner can actual operational decryption of messages begin. This is the third phase, and it is what marks the shift from cryptanalysis to practical intelligence gathering.

The TOP SECRET ULTRA "History of GYP-1," which was declassified and placed in Record Group 38 of the Archives in December 1998, and which Stinnett appears unaware of, provides a complete account of how this work was carried out in the case of the Operations Code. The code first appeared on June 1, 1939. Every few months the Japanese changed the key book, which greatly complicated the work of the codebreakers of OP-20-G; there were also regular changes in the indicator system as well as the introduction of new wrinkles, such as a codewithin-the code used for numbers and dates, and a set of auxiliary tables that added alternative definitions to the code groups when they were preceded by a special indicator group. In September 1940 an initial break was at last achieved, and a decision was made to concentrate on back traffic that had been sent using the first two key books only, as a way of maximizing the chances of breaking the indicator system and beginning to reconstruct the underlying code book; no attempt was made to catch up with current traffic yet (at this point, the Japanese were already using the fifth key book). Only two to five cryptanalysts were available to work on the project at any given time.

By October 1940, nonetheless, as Ingersoll correctly reported, the system had been "solved." But "solved" has a very different meaning to mathematical cryptanalysts than it does to the general reader, and Stinnett is badly led astray on this point. The "History of GYP-1" indeed calls particular attention to this very distinction: "Although JN-25 had been completely solved, a great deal of work remained to be done before it could be read for intelligence. The code itself had to be recovered . . . and the even more laborious job of recovering consecutive strings of additives had to be undertaken." Work in the fall of 1940 thus continued to focus on traffic that was "now over a year old," the history states. Nonetheless enough progress was made by late November 1940 that it was expected that at long last current decryption of operational traffic might begin by the new year.4

But-perhaps the most crucial fact of which Stinnett is ignorant-on December 1, 1940, an entirely new code book was introduced, superseding the previous code book. Then on February 1 and August 1, 1941, new key books were introduced as well. The crucial point is that "solving" or "breaking" a code is an ongoing process. Sometimes it is indeed a daily process, as was the case with the German Enigma traffic; every day the key changed and a huge cryptanalytic effort was required to recover the day's keys before messages could be read. Although the theory of how to break the Enigma had been determined by the first months of the war, there were still many days throughout the war when the traffic was never read. Likewise, the fact that the Japanese naval Operations Code had been "solved" in the fall of 1940-at least for year-old back traffic-does not constitute proof in the least that operational messages were being read subsequently. The entire history of cryptanalysis in World War II is one of repeated success and setbacks as enemy codes were changed or modified.

The month-by-month progress reports of OP-20-G provide even more decisive proof. These reports are clearly contemporaneous; each is date-stamped, and each is typed up in a continuous, scrolling fashion from one page to the next, so that there is no possibility of pages having been removed. They report on the complete work of OP-20-GY, the cryptanalytic section of the Navy's codebreaking bureau, and they provide precise tallies of the number of code groups and additive groups recovered each month in each system under study. They include reports on work far more sensitive than the Navy's attempts to read the Japanese naval codes, also listing progress against French diplomatic codes and Italian naval and naval attache codes. (Until only the last few years, the National Security Agency adamantly refused to declassify any documents that contained references to the fact that the United States had ever attempted to read codes of neutral or friendly nations, even during the 1920s, '30s, and '40s. The agency continues to oppose declassification of postwar documents that discuss American scrutiny of neutral nations' coded traffic.) The list of systems being worked on in these reports squares precisely with a complete list of OP-20-GY personnel assignments dated November 1941.5 The point is that we can be extremely confident that nothing has been held back in these reports. Nothing has been removed, redacted, expurgated, or censored. They were prepared as progress reports at the time the work was done; unless we grant the officers of OP-20-G the prescience in January 1940 that they would need to start concealing their work on Japanese naval codes right then and there so that FDR would be able to hide the truth two years later, we have to accept what they are plainly telling us. The progress they report on the Operations Code squares completely with the account of the "History of GYP-1"; both establish unambiguously that as a result of the code book and key book changes in the Operations Code beginning in December 1940, no current decryption at all was taking place at the time of Pearl Harbor. Indeed, they confirm that no Operations Code traffic had ever been read currently by December 7, 1941, and that no Operations Code traffic sent at any time in 1941 was read at any time before Pearl Harbor. On December 1, 1941, only 3,800 of the 30,000-plus code groups and 2,500 of the 50,000 additive had been recovered from the current system, i. e., the code book that had gone into effect December 1, 1940, and the key book that had gone into effect on August 1, 1941. (Other accounts note that most of those 3,800 code groups were the code groups that stood for numerals-which would have severely limited the ability to read any traffic, even traffic that happened to have been enciphered using the 5 percent of the key book that had been recovered.) This was simply not enough to render any messages comprehensible. A contemporaneous report which tabulates the number of messages broken out of the Operations Code in 1941-43 lists the number from 1941 as "none,"6 and this is corroborated completely by the testimony of the navy officers involved in the work and by other first-hand accounts.

When it first appeared, the Operations Code was referred to as the "five numeral system (Orange Navy)," Orange being the standard U. S. Navy codename for Japan. By August 1940 it was designated AN, and by spring 1942 it was redesignated JN-25; the latter is the name by which it is now most commonly referred. Stinnett attempts to make much of the fact that one decrypt he found in the National Archives files is of an Operations Code message which, according to intercept logs he has examined, was intercepted on November 18, 1941, but which bears the later designator "JN-25B." Above the caption "evidence of falsification" Stinnett presents a photocopy of the decrypt, and states, "The designator JN-25-B was applied to this code in late 1942 or 1943 by the Navy, but because of censorship at the time of writing of this book in 1998, the exact date cannot be determined. By pretending that this original dispatch was transmitted by Japan in JN-25-B, Navy censors have convinced many historians that this message text was not available until 1942 or later." 7 This is a mindboggling argument, as perfect an illustration of begging the question as one could ever hope to find. Because he has convinced himself that the Navy was reading the Operations Code in November 1941, Stinnett concludes that intercepting a message was tantamount to decrypting and reading it. Thus, if the decrypt of a message intercepted in November 1941 bears the designation JN-25-B, the only possible explanation is that this is a "falsification." In fact, the explanation is perfectly simple. This message, along with thousands of others transmitted by the Japanese during 1941, was indeed intercepted by the United States Navy's listening stations at the time of its transmission. It was copied, logged, passed on to the Navy codebreaking units at OP-20-G headquarters in Washington and at Hawaii, and that was that. It was riot broken and read, however, until 1946 when, the war safely won, OP-20-G decided to tidy up loose ends and go back to the accumulated raw intercepts from the pre-Pearl Harbor period. It is a standard signals-intelligence procedure to type the date of decryption (and sometimes separately the date of translation, if there was a delay) at the bottom of decrypts; thousands upon thousands of decrypts from World War II in the files of the National Archives follow this practice. At the bottom of this particular decrypt appears the words, "Navy Trans 4/24/46." Stinnett predictably concludes that this date is also "evidence of falsification," since he knows that it was actually broken before Pearl Harbor. As absurd as it sounds, that is the sum of his argument. He presents no conclusive evidence of any kind that any Operations Code messages from 1941 were actually decrypted before December 7; he merely assumes his conclusion and marches around a logical circle. The facts that texts of some of these pre-Pearl Harbor messages were published in a post-war history, or that these message were logged in (but not decrypted) by the intercept station at Hawaii, both facts that Stinnett deems highly significant, in fact mean absolutely nothing.8

In an even more fantastic excursion through the realm of negative evidence, Stinnett then points to the absence of any reference to this and other pre-Pearl Harbor decrypts in intelligence summaries issued by the Hawaii decryption station, or in material turned over the Congressional investigators, or in decrypt files released to the National Archives in 1979. Clearly this was all part of the cover up, he argues, done "deliberately to conceal American success in decoding Japanese naval communications." Similar paranoid reasoning appears throughout the book. Why wasn't Agnes Driscoll, who worked on the Operations Code, awarded a citation for her work? Simple: "Of course the Navy could not publicly credit her with the solution. Not to do so was a deliberate decision devised to mask America's success in learning Japanese military secrets." As alien as it may to be those who minds operate in these channels, it is at least worth considering the possibility that sometimes the reason no record of an event can be found is that the event in question never happened.

At times it seems Stinnett forgets what he is trying to prove. At one point he makes much of the supposed fact that the Hawaii unit was working on the Operations Code in 1940 and 1941, despite post-war testimony by its officer in charge, Joseph Rochefort,9 and by the Pacific Fleet's intelligence officer, Edwin Layton, that the job was taken away from Hawaii and given to Washington and Cavite. His evidence for this is that Hawaii was provided by Washington with "solutions" of the code in April 1941 and fall of 1941 and that in November 1940 "special IBM sorting equipment for use in decrypting the 5-Num [i.e., Operations code" was ordered for Hawaii. But this, too is nonsense; IBM machines were used for decrypting all enciphered codes, including German diplomatic codes, Russian diplomatic and KGB codes, Japanese army codes, and many other Japanese navy codes besides the Operations Code. There was no such thing as "special" IBM sorting equipment designed for just the Operations Code.10 (Hawaii had been assigned what proved to be the dead-end task of the attacking the Japanese navy "AD" code, apparently a little-used flag officer's and administrative system; according to Rochefort's and Layton's accounts, fully corroborated in the month-by-month progress reports, Hawaii had no success whatsoever in breaking this system. Nor were the other high-level Japanese Navy codes being studied by Washington, Hawaii, and Cavite broken.) Moreover, as Layton fully documents in his memoirs, And I Was There (and as Stinnett elsewhere acknowledges himself), the plan up until July 1941 had been to turn over the Operations Code to Hawaii; it was only a last minute change that led to the job being retained in Washington, with Cavite assisting in the additive recovery task. Elsewhere Stinnett confuses the intelligence that the Hawaii station was deriving from the process known as traffic analysis with actual decryption. The Navy radio intercept station at Hawaii was not only routinely intercepting all Operations Code traffic it could pick up, it was also taking radio bearings on the signals and studying call signs and other "externals" of the messages for clues about the location and movement of Japanese naval units. But the fact that references to radio signals transmitted in the Operations Code appear in Rochefort's intelligence reports says absolutely nothing about whether Hawaii was attempting to break and read the contents of the messages themselves; in fact all the evidence is in perfect agreement that Hawaii was not.

But what is Stinnett's point in any case? He indulges in much heavy breathing about the "cover-up" of Hawaii's 1941 work on the Operations Code. But if Hawaii was doing this work, and was breaking the code, then how can he possibly explain the idea that FDR knew but the local naval commanders did not? Layton and Rochefort established an extremely close working relationship, and although Rochefort nominally was under the command of the commandant of the Fourteenth Naval District, a deskborne administrative admiral, in fact he saw his job as providing intelligence directly to the commander of the Pacific Fleet. Layton describes in detail how this cooperation worked, with Rochefort filing intelligence reports directly to him and the two speaking by phone often several times a day. To assert that Hawaii was working on the Operations Code does perforce make Rochefort and Layton and the Navy out to be liars, and suggest somebody was covering up something, but doesn't this rather undermine the conspiratorial thesis that FDR somehow concealed the news from local commanders?

About the only real goods Stinnett delivers in 400 pages is a hand-written note by Captain Joseph Redman, assistant director of naval communications, ordering that a reference to cooperation with the British on the Operations Code in spring 1941 be removed from the files. Stinnett says this was because it would have disclosed the "successful British effort aimed at penetrating the 5-Num system." But again, Stinnett is playing with half understood facts. As a result of the establishment of the highly secret U. S.-British cryptanalytic cooperation that began in February 1941, the British codebreaking unit at Singapore at once began sharing results on the Operations Code with the U. S. Navy unit at Cavite. But again, all of the contemporaneous documents, and later testimony by British and American official involved, agree that while the cooperation did lead to a pooling of additive recoveries (indeed, the British and the Americans had made about equal progress by the spring of 1941, though in different parts of the additive book, so by pooling their results they quickly doubled the number of additive recoveries made), it simply did not change the final outcome. The British were hampered by a terrible shortage of Japanese linguists; indeed in February 1941 they had proposed turning over responsibility for all work on Japanese naval codes in the Far East to the Americans if the Americans could only supply the extra linguistic help required. Even OP-20-G by the fall of 1941 could spare only about a total of twenty people to work on the Operations Code. And in any case the key-book change in August 1, 1941 (of which Stinnett is totally unaware) proved to be an unrecoverable blow, because it meant starting from scratch. There is also, of course, a perfectly simple reason for Redman's order, namely the urgent desire to keep the very fact of U. S.-British cooperation secret. At the time there is no doubt it would have been political dynamite in the hands of American isolationists if it were known that the United States was sharing its most intimate secrets with perfidious Albion; even after the war, the fact and terms of U. S.-British secret agreements on signals intelligence cooperation were guarded with the strictest classification well into the 1980s. It was only in 1996 that the full, unexpurgated text of the 1943 wartime "BRUSA" agreement on signals intelligence collaboration was declassified.

After Pearl Harbor, powered by a huge surge of manpower and galvanized by the urgent needs of wartime, Rochefort's group in Hawaii and OP-20-G in Washington joined forces on B-25 and were able to begin current decryption on March 18, 1942. The current reading of JN-25 was what led to the stunning intelligence coup by U. S. forces at the Battle of Midway in June; broken JN-25 messages foretold of the Japanese plan to ambush the U. S. carriers, allowing Admiral Nimitz to get there first and ambush the ambushers.

However, all of the documentary evidence-and it is considerable-debunks this latest attempt to grasp at conspiratorial straws about Pearl Harbor. The Japanese naval codes that might have given specific or at least strongly circumstantial warnings of the Japanese plans for the strike against the American fleet on December 7 simply were not being read during 1941. That other indications of Japanese intentions were ignored is undeniable, the most notorious lapse being the failure of operators of the U. S. radar station in Hawaii to see anything worrisome in the huge swarm of aircraft that appeared on their screens (the officer in charge incredibly assumed they were a flight of American Army bombers coming from the mainland.) That Roosevelt wanted to get America into the. war, and that it would take a military incident to do it, is undeniable. That America knew by December 6 that Japan was preparing for war and would no doubt attack is undeniable. But this new attempt to weave together a conspiracy by FDR to conceal hard intelligence that he obtained-but which somehow the Navy did not just does not hold water. It would be a mercy if those who are so fascinated by this subject would in the future read the archival materials that are available for all to examine before they commit their theories to paper.

BIOGRAPHICAL SKETCH

Stephen Budiansky is an author and science journalist. His history of Allied codebreaking in World War II, Battle of Wits, will be published by The Free Press in October 2000.

1See my forthcoming article in Intelligence and National Security, "The difficult beginnings of U. S.-British codebreaking cooperation".

2Gore Vidal, "Candid in Camera," The Nation, Sept. 27, 1999.

3The internal history of the Operations Code and its solution is titled "History of GYP-1," and is found in the Crane Files, Record Group 38, CNSG 5750/202, National Archives at College Park. The month-by-month progress reports are located in the file "OP-20-GY," Crane Files, Record Group 38, CNSG 5750/198.

a "History of GYP-1," Crane Files, Record Group 38, CNSG 5750/202, pp. 14-18.

5File #62 War Diary File on History and Functions of OP-20-G Offices, report dated November 11, 1941, Crane Files, Record Group 38, CNSG 5750/201.

6"The Activities and Accomplishments of GY-1 During 1941; 1942 and 1943," Crane Files, Records Group 38, CNSG 5750/197.

7When the designation JN-25 was adopted, it was applied by the Navy retroactively from that point on to the all earlier traffic sent in the Operations Code, so as a matter of course an Operations Code message from 1941 decrypted at a later dagte would be designated JN-25B. The "B" refers to the edition of the code book being used. The first code book, in use from June 1, 1939, to November 30, 1940, was originally designated Able; the second code book, in force from December 1, 1940, to May 27, 1942 was Baker; this nomenclature was carried directly over, and thus JN-25B referred to all messages sent using the Baker code book.

8A complete set of these 1945 and 1946 decrypts is available in Record Group 38: "Pre-Pearl Harbor Japanese Naval Dispatches," Crane Files, CNSG 5830/115. Stinnett appears to be unaware of them, even though most were also published in 1994 by the National Security Agency in an unclassified monograph, "Pearl Harbor Revisited: United States Navy Communications Intelligence, 1924-1941." This monograph also contains a summary of the daily intelligence reports issued by the Hawaii unit to Admiral Kimmel, the commander of the Pacific Fleet.

9See Rochefort's oral history, U. S. Naval Institute.

10A number of special purpose attachments to IBM machines were developed by Arlington Hall to aid in the attack on certain cryptanalytic problems, and later in the war the Navy took the lead in constructing a number of special purpose "Rapid Analytical Machines" that proved especially valuable in breaking the Soviet codes and the Japanese naval attache codes. But in 1940 and 1941 the only mechanical devices in routine use against the Japanese navy codes were standard IBM sorters and tabulators.

Stephen Budiansky

ADDRESS: Black Sheep Farm, 14605 Chapel Lane, Leesburg VA 20176 USA.

Copyright Cryptologia Apr 2000
Provided by ProQuest Information and Learning Company. All rights Reserved