TheCodeBreakers (0)

Some of the things you will learn in THE CODEBREAKERS 
 
•  How secret   Japanese  messages were decoded in Washington hours before Pearl   Harbor . 
•  How  German codebreakers helped usher in the Russian Revolution . 
•  How John F. Kennedy  escaped capture  in the Pacific because the Japanese failed to solve a 
simple cipher . 
•  How codebreaking determined a presidential election,  convicted  an underworld  syndicate 
head, won the  battle of Midway , led to cruel Allied defeats in North Africa , and broke up a vast 
Nazi spy ring. 
•  How one American became the world's most  famous  codebreaker, and another became the 
world's greatest . 
•  How codes and codebreakers operate today   within  the secret agencies of the U.S. and Russia . 
•  And incredibly much more. 
"For many evenings of gripping  reading , no better choice can be made than  this book." 
— Christian   Science   Monitor  
 
THE 
Codebreakers 
 
The Story of Secret Writing 
 
By DAVID KAHN  
( abridged by the author ) 
 
A SIGNET BOOK from 
NEW AMERICAN LIBRARV
TIMES MIRROR  
 
Copyright © 1967, 1973 by David Kahn 
All rights reserved. 
No part of this book may be reproduced or transmitted 
in any form or by any means , electronic or  mechanical , 
including  photocopying, recording or by any information 
storage  and retrieval system,  without permission  in writing 
from the publisher . For information address 
The Macmillan Company, 866 Third Avenue, New York , 
New York 10022. 
Library of Congress Catalog Card Number: 63-16109 
Crown copyright is acknowledged for the following illustrations 
from Great Britain 's Public  Record  Office: 
S.P. 53/18, no. 55, the Phelippes forgery, 
and P.R.O. 31/11/11, the Bergenroth reconstruction . 
Published by arrangement with The Macmillan Company 
FIRST PRINTING SECOND PRINTING THIRD PRINTING  FOURTH PRINTING FIFTH PRINTING  SIXTH PRINTING SEVENTH PRINTING 
EIGHTH PRINTING NINTH PRINTING  TENTH  PRINTING 
SIGNET TRADEMARK: REG. TJ.S. PAT. OFF. AND  FOREIGN COUNTRIES 
REGISTERED TRADEMARK---MARCA REGISTBADA 
HECHO EN CHICAGO , U.S.A. 
SIGNET, SIGNET CLASSICS, SIGNETTE,  MENTOR AND PLUME BOOKS  
are published by The New American Library, Inc., 
1301 Avenue of the Americas , New York, New York 10019 
FIRST PRINTING,  FEBRUARY , 1973 
PRINTED IN THE UNITED STATES OF AMERICA 
 
To my Parents  
and my Grandmother 
 
Contents 
A Note  on the Abridged  Version
Preface
A Few Words
1.  One Day of  Magic : I
2.  One Day of Magic: II
3.  The First 3,000  Years
4.  The Rise  of the  West
5.  On the  Origin of a  Species
6.  The Era of the Black Chambers
7.  The Contribution of the Dilettantes
8.   Room  40
9.  A War of Intercepts
10.  Two Americans
11.  Secrecy for  Sale
12.   Duel in the Ether: I
13.  Duel in the Ether: II
14.  Censors, Scramblers, and Spies
15.  The Scrutable Orientals
16.  PYCCKAJI Kranrojioras
17.  N.S.A.
18.  Heterogeneous Impulses
19.  Ciphers in the Past  Tense
20.  The  Anatomy  of Cryptology
Suggestions for Further  Reading 
Index
A Note on the Abridged Version 
 
MANY PEOPLE have  urged  me to put out a paperback edition  of The Codebreakers.  Here it is. 
It comprises about a third of the  original . This was as big as the publishers and I could  make it and  still  
keep the  price  within  reason . 
In cutting the book, I retained  mainly stories  about how codebreaking has  affected history, particularly  
in World War II, and major   names and stages in the history of cryptology. I eliminated all source notes and 
most of the technical matter , as well as material peripheral to strict codebreaking  such  as biographies, the 
invention  of secondary cipher systems, and miscellaneous uses of various systems. 
I had no space  for new material, but I did correct the errors reported to me and updated a few items . 
The chapters  have been slightly rearranged. 
Readers wanting to  know more about a specific point should consult the text and notes of the original. 
If any reader wishes to  offer any corrections or to tell  me of his own  experiences in this  field , I would 
be very grateful if he would send   them  to me. 
—D.K. 
Windsor Gate  
Great Neck, New York 
 
Preface 
 
CODEBREAKING is the most important  form of secret intelligence in the world today. It produces much more 
and much more trustworthy information than spies, and this intelligence exerts great influence upon the 
policies of governments. Yet it has  never  had a chronicler. 
It badly  needs  one. It has been  estimated that cryptanalysis saved a year of war in the Pacific, yet the 
histories give it but  passing   mention . Churchill 's great history of World War II has been cleaned of every 
single reference  to Allied communications intelligence except one (and that based on the American Pearl 
Harbor investigation), although Britain  thought  it vital enough to assign 30,000 people to the work . The 
intelligence history of World War II has never been written. All this gives a distorted view of why things 
happened . Furthermore , cryptology itself can  benefit , like  other spheres of human endeavor, from knowing 
its major trends , its great men, its errors made and lessons learned. 
I have tried in this book to write a serious  history of cryptology. It is primarily a report to the public on 
the important  role that cryptology has played, but it may also orient cryptology with regard to its past and 
alert historians to the sub rosa influence of cryptanalysis. The book seeks to  cover the entire history of 
cryptology. My  goal  has been twofold: to narrate the  development  of the various methods of  making and 
breaking codes and ciphers, and to tell how  these methods have affected men. 
When I began this book, I, like other well-informed amateurs, knew about all that had been published 
on the history of cryptology in books on the  subject . How little we really knew! Neither we nor any 
professionals realized that many  valuable   articles lurked in scholarly journals , or had induced any 
cryptanalysts to tell their stories for publication, or had tapped the vast treasuries of  documentary  material, 
or had tried to take a long view and ask some  questions that now appear  basic . I believe it to be true that, 
from the point of view of the material previously published in books on cryptology, what is new in this 
book is 85 to 90 per cent . 
Yet it is not exhaustive. A foolish secrecy still clothes much of World War II cryptology—though I 
believe the outlines of the achievements are  known —and to tell just that story in full would  require a book 
the size  of this.  Even  in, say, the  18th century , the unexplored manuscript material is very great. 
Nor is this a textbook . I have sketched a few methods of  solution . For some readers even this will be too 
much; them I advise skip this material. They will not have a full understanding  of what is going  on, but that 
will not cripple their comprehension of the stories. For readers who want more detail on these methods, I 
recommend, in the  rear  of this book, some other  works  and  membership in the American Cryptogram 
Association . 
In my writing, I have tried to adhere to two principles. One was to use  primary   sources as much as 
possible. Often it could not be  done  any other way,  since   nothing  had been published on a  particular matter. 
The other principle was to try to make certain that I did not give cryptology sole and total credit for 
winning  a battle or making possible a diplomatic coup or whatever happened if, as was usual, other factors 
played a role. Narratives which make it appear as if every event in history turned upon the subject under 
discussion are not history but journalism. They are especially prevalent in spy stories, and cryptology is not 
immune . The only other book- length  attempt to survey the history of cryptology, the late Fletcher Pratt's 
Secret and Urgent , published in 1939, suffers from a severe  case of this special pleading. Pratt writes 
thrillingly— perhaps  for that very reason—but his failure  to  consider the other factors, together with his 
errors and omissions, his false generalizations based on no  evidence , and his unfortunate predilection for 
inventing facts vitiate his work as any kind of a history. ( Finding  this out was disillusioning, for it was this 
book, borrowed from the Great Neck Library, that interested me in cryptology.) I think that although trying 
to  balance the story with the other factors may detract a little from the immediate thrill, it charges it with 
authenticity and  hence makes for long-lasting  interest : for this is how things really happened. 
In the same  vein, I have not made up any conversations, and my speculations about things not a matter 
of record have been marked as such in the notes in the full-length version. I have documented all important 
facts, except that in a few cases I have had to  respect the wishes of my sources for anonymity. 
The original publisher submitted the manuscript to the  Department  of  Defense  on  March  4, 1966, which 
requested three  minor deletions—to all of which I acceded—before releasing the manuscript for 
publication. 
DAVID KAHN 
Windsor Gate 
Great Neck, New York 
Paris  
A Few Words 
 
 
EVERY TRADE has its  vocabulary . That of cryptology is simple, but even so a familiarity with its terms  
facilitates understanding. A glossary may also  serve as a handy  reference. The definitions in this one are 
informal  and ostensive.  Exceptions  are ignored and the  host  of minor terms are not defined—the text covers 
these when they  come  up. 
The plaintext is the message that will be put into secret form. Usually the plaintext is in the  native  
tongue of the communicators. The message may be hidden  in two basic  ways . The methods of 
steganography conceal the very existence of the message.  Among them are invisible inks and microdots 
and arrangements in which, for example, the first letter  of each word in an apparently innocuous text spells 
out the  real  message. (When steganography is applied to  electrical communications, such as a method that 
transmits a long  radio  message in a single short spurt, it is called transmission security .) The methods of 
cryptography, on the other hand , do not conceal the presence of a secret message but  render it 
unintelligible to outsiders by various transformations of the plaintext. 
Two basic transformations  exist . In transposition, the letters  of the plaintext are jumbled; their normal 
order  is disarranged. To  shuffle  secret into ETCRSE is a transposition. In  substitution , the letters of the 
plaintext are replaced by other letters, or by numbers or  symbols . Thus secret might become 19 5 3 18 5 20, 
or XIWOXY in a more complicated system. In transposition, the letters retain  their identities— the two e's of 
secret are still  present in ETCRSE—but they  lose their positions , while in substitution the letters retain their 
positions but lose their identities. Transposition and substitution may be combined. 
Substitution systems are much more  diverse and important than transposition systems. They rest on the 
concept  of the cipher alphabet . This is the list of equivalents used to transform the plaintext into the secret 
form. A sample cipher alphabet might be: 
 
plaintext letters  abcdefghijklm 
cipher letters  
LBQACSRDTOFVM 
plaintext letters nopqrstuvwxyz 
cipher letters 
HWIJXGKYUNZEP 
 
This graphically indicates that the letters of the plaintext are to be replaced by the cipher letters beneath 
them, and vice versa. Thus, enemy would become  CHCME , and swc would  reduce  to foe. A set of such 
correspondences is still called a "cipher alphabet" if the plaintext letters are in mixed order, or even if they 
are missing , because cipher letters always imply plaintext letters. 
Sometimes such an alphabet will provide multiple substitutes for a letter. Thus plaintext e, for 
example, instead of always being replaced by, say, 16, will be replaced by any one of the  figures 16, 74, 35, 
21. These alternates are called homophones. Sometimes a cipher alphabet will include symbols that mean  
nothing and are intended to confuse interceptors; these are called nulls. 
As long as only one cipher alphabet is in use, as above , the system is called monoalpbabetic. When, 
however , two or more cipher alphabets are employed in some kind of prearranged pattern, the system 
becomes polyalphabetic. A simple form of polyalphabetic substitution would be to add another cipher 
alphabet under the one  given  above and then to use the two in  rotation , the first alphabet for the first 
plaintext letter, the second for the second, the first again for the third plaintext letter, the second for the 
fourth, and so on. Modern cipher machines produce polyalphabetic ciphers that employ   millions  of cipher 
alphabets. 
Among the systems of substitution,  code  is distinguished from cipher. A code  consists  of thousands of 
words,  phrases , letters, and syllables with the codewords or code-numbers (or, more generally, the 
codegroups) that replace these plaintext elements . 
 
plaintext  
codeword 
emplacing   DVAP 
employ  
DVBO 
en-  
DVCN 
enable  
DVDM 
enabled 
DVEL 
enabled to 
DVFK 
 
This means, of course , that DVDM replaces enable. If the plaintext and the code elements both  run in 
alphabetical or  numerical  order, as above, the code is a one-part code, because a single book serves for 
both en- and decoding. If, however, the code equivalents stand in mixed order opposite their plaintext 
elements, like this 
 
Plaintext codenumber 
shield (for)  51648 
shielded 
07510 
shielding 
10983 
shift (s) 
43144 
ship  
35732 
ships 
10762 
 
the code is a two-part code, because a second section , in which the code elements are in  regular  order, is 
required  for decoding: 
 
codenumber  
plaintext 
10980 was 
not 
10981 spontaneous 
(ly) 
 
10983 
shielding  
10986 
April 13  
10988 
withdrawn from  
10990 
acknowledge 
 
In a sense , a code comprises a gigantic cipher alphabet, in which the basic plaintext  unit  is the word or 
the phrase ; syllables and letters are supplied mainly to  spell out words not present in the code. In ciphers, 
on the other hand, the basic unit is the letter, sometimes the letter- pair  (digraph or bigram), very rarely 
larger groups of letters (polygrams). The substitution and transposition systems illustrated above are 
ciphers.  There  is no  sharp theoretical dividing line  between  codes and ciphers; the latter  shade into the 
former as they  grow larger. But in modern practice the  differences  are usually quite marked. Sometimes the 
two are distinguished by saying that ciphers operate on plaintext  units  of regular length (all single letters or 
all groups of, say, three letters), whereas codes operate on plaintext groups of variable length (words, 
phrases, individual  letters, etc.). A more penetrating and useful distinction is that code operates on 
linguistic entities, dividing its raw material into meaningful elements like words and syllables, where as 
cipher does not—cipher will split the t from the h in the, for example. 
For 450 years, from about  1400  to about 1850, a system that was  half  a code and half a cipher 
dominated cryptography. It usually had a separate cipher alphabet with homophones and a codelike list of 
names, words, and syllables. This list, originally just of names, gave  the system its name: nomenclator. 
Even though late in its life some nomenclators grew larger than some modern codes, such systems are still 
called "nomenclators" if they  fall within this historical period . An odd characteristic is that nomenclators 
were always written on large folded sheets of  paper , whereas modern codes are almost invariably in book 
or  booklet  form. The commercial code is a code used in business primarily to save on  cable tolls; though 
some are compiled for private  firms , many others  are  sold  to the public and therefore  provide no real 
secrecy. 
Most ciphers employ a key, which specifies such things as the arrangement of letters within a cipher 
alphabet, or the pattern of shuffling in a transposition, or the settings on a cipher  machine . If a word or 
phrase or number serves as the key, it is naturally called the keyword or keyphrase or keynumber. Keys 
exist within a general system and control that system's variable elements. For example, if a polyalphabetic 
cipher provides 26 cipher alphabets, a keyword might  define the half dozen or so that are to be used in a 
particular message. 
Codewords or codenumbers can be subjected to transposition or substitution just like any other group 
of letters or numbers—the transforming  processes do not ask that the texts given to them be intelligible. 
Code that has not yet undergone such a  process —called superencipherment —or which has been 
deciphered from it is called placode, a shortening of " plain  code." Code that has been transformed is called 
encicode, from "enciphered code." 
To  pass  a plaintext  through these transformations is to encipher or encode it, as the case may be. What 
comes out of the transformation is the ciphertext or the codetext. The final secret message,  wrapped  up 
and sent , is the cryptogram. (The term  "ciphertext" emphasizes the result  of encipherment more, while 
"cryptogram" emphasizes the  fact of transmission more; it is analogous to " telegram .") To decipher or 
decode is for the persons legitimately possessing the key and system to  reverse the transformations and 
bare  the original message. It  contrasts with cryptanalyze, in which persons who do not possess the key or 
system— a third  party , the "enemy"—break down or solve the cryptogram. The  difference  is, of course, 
crucial . Before about 1920, when the word cryptanalysis was coined to mean the methods of breaking 
codes and ciphers, "decipher" and "decode" served in both senses (and  occasionally still do), and in 
quotations  where they are used in the sense of solve, they are retained if they will not confuse. Sometimes 
cryptanalysis is called codebreaking; this includes solving ciphers. The original intelligible text that 
emerges from either decipherment or cryptanalysis is again called plaintext. Messages sent without 
encipherment are cleartext or in clear , though they are sometimes called in plain language . 
Cryptology is the science that embraces cryptography and cryptanalysis, but the term "cryptology" 
sometimes loosely designates the entire  dual  field of both rendering  signals  secure and extracting 
information from them. This broader field has grown to include many new  areas ; it encompasses, for 
example, means to deprive the enemy of information obtainable by  studying  the  traffic   patterns  of radio 
messages, and means of obtaining information from  radar  emissions. An  outline of this larger field, with its 
opposing parts placed opposite one another, and with a few of the methods of each part given in 
parentheses, would be: 
 
SIGNAL   SECURITY                                            SIGNAL INTELLIGENCE 
 
Communication Security 
 
Communication Intelligence
Steganography (invisible inks, op n 
e
Interception and Direction-Finding 
codes, messages in hollow  heels) 
and Transmission Security (spurt 
radio systems)  
Traffic Security  ( call - sign   changes , 
Traffic   Analysis    (direction-finding 
dummy messages, radio silence) 
fixes , message-flow  studies , radio- 
fingerprinting)  
 
Cryptography   (codes  and  ciphers,  
Cryptanalysis 
ciphony, cifax) 
 
Electronic Security 
 
Electronic Intelligence
Emission  Security ( shifting of ra- 
Electronic Reconnaissance (e v
a es- 
dar frequencies)   
 
dropping on radar emissions) 
Counter -Countermeasures (" look - 
Countermeasures (jamming, false 
ing-through" jammed radar)  radar echoes) 
 
This book employs certain typographic conventions for simplicity and economy . Plaintext is always set 
lower case; when it occurs in the running text (as opposed to its occurrence in the diagrams ), it is also in 
italics . Cipher-text or codetext is set in SMALL CAPS  in the text, keys in LARGE CAPS. They are 
distinguished in the diagrams by labels. Cleartext and translations of foreign-language plaintext are in 
roman  within  quotation   marks . The sound  of a letter or  syllable or word, as distinguished from its written 
form, is placed within diagonals, according to the  convention widely followed in linguistics; thus /t/ refers 
to the unvoiced stop normally represented by that letter and not to the graphic   symbol  t. 
D. K. 
 
 
1. One Day of Magic:  I 
 
AT 1:28 on the morning of December 7, 1941, the big ear of the Navy's 
radio station on Bainbridge Island near Seattle trembled to vibrations in 
the ether. A message was coming through on the Tokyo -Washington 
circuit . It was addressed to the Japanese embassy, and Bainbridge 
reached up and snared it as it flashed overhead. The message was short, 
and its radiotelegraph transmission took only nine minutes. Bainbridge 
had it all by 1:37. 
The station's personnel punched the intercepted message on a 
teletype tape, dialed a number on the teletypewriter exchange , and when 
the connection had been made, fed the tape into a mechanical 
transmitter that gobbled it up at 60 words per minute . 
The intercept reappeared on a page-printer in Room 1649 of the Navy 
Department building on Constitution Avenue in Washington, D.C. What 
went on in this room, tucked for security's sake at the end of the first 
deck 's sixth wing , was one of the most closely guarded secrets of the 
American government . For it was in here—and in a similar War 
Department room in the Munitions Building next door —that the United 
States peered into the most confidential thoughts and plans of its 
possible enemies by shredding the coded wrappings of their dispatches. 
Room 1649 housed OP-20-GY, the cryptanalytic section of the Navy's 
cryptologic organization, OP-20-G. The page-printer stood beside the 
desk of the GY watch officer. It rapped out the intercept in an original and 
a carbon copy on yellow and pink teletype paper just like news on a city 
room wireservice ticker. The watch officer, Lieutenant (j.g.) Francis M. 
Brotherhood, U.S.N.R., a curly-haired, brown - eyed six- footer , saw 
immediately from indicators that the message bore for the guidance of 
Japanese code clerks that it was in the top Japanese cryptographic 
system. 
This was an extremely complicated machine cipher which American 
cryptanalysts called  PURPLE . Led by William F. Friedman , Chief  
Cryptanalyst of the Army Signal Corps , a team of codebreakers had 
solved Japan 's enciphered dispatches, deduced the nature of the 
mechanism that would effect those letter transformations, and 
painstakingly built up an apparatus that cryptographically duplicated 
the Japanese machine. The Signal Corps had then constructed several  
additional PURPLE machines, using a hodgepodge of manufactured parts, 
and had given one to the Navy. Its three components rested now on a 
table in Room 1649: an electric typewriter for input; the cryptographic 
assembly proper , consisting of a plugboard, four electric coding rings , 
and associated wires and switches, set on a wooden frame ; and a 
printing unit for output. To this precious contraption, worth quite 
literally more than its weight in gold , Brotherhood carried the intercept. 
He flicked the switches to the key of December 7. This was a 
rearrangement, according to a pattern ascertained months ago, of the 
key of December 1, which OP-20-QY had recovered. Brotherhood typed 
out the coded message. Electric impulses raced through the maze of 
wires, reversing the intricate enciphering process. In a few minutes, he 
had the plaintext before him. 
It was in Japanese. Brotherhood had taken some of the orientation 
courses in that difficult language that the Navy gave to assist its 
cryptanalysts. He was in no sense a translator , however, and none was 
on duty next door in OP-20-GZ, the translating section. He put a red 
priority sticker on the decode and hand-carried it to the Signal 
Intelligence Service , the Army counterpart of OP-20-O, where he knew 
that a translator was on overnight duty. Leaving it there, he returned to 
OP-20-G. By now it was after 5 a.m. in Washington—the message having  
lost three hours as it passed through three time zones in crossing the 
continent . 
The S.I.S translator rendered the Japanse as: "Will the Ambassador 
please submit to the United States Government (if possible to the 
Secretary of State) our reply to the United States at 1:00 p.m. on the 7th, 
your time." The —"reply" referred to had been transmitted by Tokyo in 14 
parts over the past 18½  hours, and Brotherhood had only recently 
decrypted the 14th part on the PURPLE machine. It had come out in the 
English in which Tokyo had framed it, and its ominous final sentence 
read: "The Japanese Government regrets to have to notify hereby the 
American Government that in view of the attitude of the American 
Government it cannot but consider that it is impossible to reach an 
agreement through further negotiations ." Brotherhood had set it by for 
distribution early in the morning. 
The translation of the message directing delivery at one o' clock had 
not yet come back from S.I.S. when Brotherhood was relieved at 7 a.m., 
and he told his relief, Lieutenant (j.g.) Alfred V. Pering, about it. Half an 
hour later , Lieutenant Commander Alwin D. Kramer , the Japanese-
language expert who headed GZ and delivered the intercepts, arrived . He 
saw at once that the all-important conclusion of the long Japanese 
diplomatic note had come in since he had distributed the 13 previous  
parts the night before. He prepared a smooth copy from the rough decode 
and had his clerical assistant, Chief Yeoman H. L. Bryant , type up the 
usual 14 copies. Twelve of these were distributed by Kramer and his 
opposite number in S.I.S. to the President , the secretaries of State, War, 
and Navy, and a handful of top-ranking Army and Navy officers. The two 
others were file copies. This decode was part of a whole series of 
Japanese intercepts, which had long ago been given a collective 
codename, partly for security, partly for ease of reference, by a previous 
director of naval intelligence, Rear Admiral Walter S. Anderson . Inspired, 
no doubt , by the mysterious daily production of the information and by 
the aura of sorcery and the occult that has always enveloped cryptology, 
he called it MAGIC. 
When Bryant had finished , Kramer sent S.I.S. its seven copies, and at 
8 o'clock took a copy to his superior , Captain Arthur H. McCollum , head 
of the Far Eastern Section of the Office of Naval Intelligence. 
 
From: Tokyo 
To:   Washington 
December 7, 1941 
Purple (Urgent - Very Important) 
#907. 
To be handled in goverment code. 
Re: my #902a. 
Will the Ambaagador please submit to the United States 
Government (If possible to the Secretary of State) our reply to 
the United States at 1:00 p.m. on the 7th, your time. 
 
a - JD-1:7143 - text of Japanese reply. 
MAGIC'S solution of the Japanese one o'clock delivery message 
 
He then busied himself in his office, working on intercepted traffic, 
until 9:30, when he left to deliver the 14th part of Tokyo's reply to 
Admiral Harold F. Stark , the Chief of Naval Operations, to the White 
House, and to Frank Knox, the Secretary of the Navy. Knox was meeting  
at 10 a.m. that Sunday morning in the State Department with Secretary 
of War Henry L. Stimson and Secretary of State Cordell Hull to discuss  
the critical nature of the American negotiations with Japan, which, they 
knew from the previous 13 parts, had virtually reached an impasse. 
Kramer returned to his office about 10:20, where the translation of the 
message referring to the one o'clock delivery had arrived from S.I.S. while 
he was on his rounds . 
Its import crashed in upon him at once. It called for the rupture of 
Japan's negotiations with the United States by a certain deadline . The 
hour set for the Japanese ambassadors to deliver the notification—1 p.m. 
on a Sunday—was highly unusual . And, as Kramer had quickly 
ascertained by drawing a navigator's time circle , 1 p.m. in Washington 
meant 7:30 a.m. in Hawaii and a couple of hours before dawn in the 
tense Far East around Malaya, which Japan had been threatening with 
ships and troops. 
Kramer immediately directed Bryant to insert the one o'clock message 
into the reddish-brown looseleaf cardboard folders in which the MAGIC 
intercepts were bound . He included several other intercepts, adding one 
at the last minute, then slipped the folders into the leather briefcases, 
zipped these shut , and snapped their padlocks. Within ten minutes he 
was on his way. 
He went first to Admiral Stark's office, where a conference was in 
session, and indicated to McCollum, who took the intercept from him, 
the nature of the message and the significance of its timing. McCollum 
grasped it at once and disappeared into Stark's office. Kramer wheeled 
and hurried down the passageway. He emerged from the Navy 
Department building and turned right on Constitution Avenue, heading  
for the meeting in the State Department four blocks away . The urgency of 
the situation washed over him again, and he began to move on the 
double . 
 
This moment, with Kramer running through the empty streets of 
Washington bearing his crucial intercept, an hour before sleepy code 
clerks at the Japanese embassy had even deciphered it and an hour 
before the Japanese planes roared off the carrier flight decks on their 
treacherous mission , is perhaps the finest hour in the history of 
cryptology. Kramer ran while an unconcerned nation slept late, ignored 
aggression in the hope that it would go away, begged the hollow gods of 
isolationism for peace , and refused to entertain—except humorously—the 
possibility that the little yellow men of Japan would dare attack the 
mighty United States. The American cryptanalytic organization swept  
through this miasma of apathy to reach a peak of alertness and 
accomplishment unmatched on that day of infamy by any other agency  
in the United States. That is its great achievement , and its glory. 
Kramer's sprint symbolizes it. 
Why, then, did it not prevent Pearl Harbor? Because Japan never sent 
any message saying anything like "We will attack Pearl Harbor." It was 
therefore impossible for the cryptanalysts to solve one. Messages had 
been intercepted and read in plenty dealing with Japanese interest in 
warship movements into and out of Pearl Harbor, but these were 
evaluated by responsible intelligence officers as on a par with the many 
messages dealing with American warships in other ports and the Panama  
Canal . The causes of the Pearl Harbor disaster are many and complex , 
but no one has ever laid any of whatever blame there may be at the doors  
of OP-20-G or S.I.S. On the contrary, the Congressional committee that 
investigated the attack praised them for fulfilling their duty in a manner  
that "merits the highest commendation." 
As the climax of war rushed near, the two agencies— together the 
most efficient and successful codebreaking organization that had ever 
existed—scaled heights of accomplishment greater than any they had 
ever achieved. The Congressional committee, seeking the responsibility 
for the disaster, exposed their activity on almost a minute-by-minute 
basis . For the first time in history, it photographed in fine-grained detail 
the operation of a modern code-breaking organization at a moment of 
crisis . This is that film . It depicts OP-20-G and S.I.S. in the 24 hours 
preceding the Pearl Harbor attack, with the events of the past as 
prologue. It is the story of one day of MAGIC. 
 
The two American cryptanalytic agencies had not sprung full-blown 
into being like Athena from the brow of Zeus. The Navy had been solving 
at least the simpler Japanese diplomatic and naval codes in Rooms 1649 
and 2646 on the "deck" above since the 1920s . The Army's 
cryptanalytical work during the 1920s was centered in the so-called 
American Black Chamber under Herbert O. Yardley, who had organized 
it as a cryptologic section of military intelligence in World War I. It was 
maintained in secrecy in New York jointly by the War and State 
departments, and perhaps its greatest achievement was its 1920 solution 
of Japanese diplomatic codes. At the same time, the Army's cryptologic 
research and code-compiling functions were handled by William 
Friedman, then as later a civilian employee of the Signal Corps. In 1929, 
Henry L. Stimson, then Secretary of State, withdrew State Department 
support from the Black Chamber on ethical grounds , dissolving it. The 
Army decided to consolidate and enlarge its codemaking and 
codebreaking activities . Accordingly, it created the Signal Intelligence 
Service, with Friedman as chief, and, in 1930, hired three junior 
cryptanalysts and two clerks. 
The following year, a Japanese general suddenly occupied Manchuria 
and set up a puppet Manchu emperor , and the government of the island 
empire of Nippon fell into the hands of the militarists. Their avarice for 
power , their desire to enrich their have-not nation, their hatred for white 
Occidental civilization, started them on a decade-long march of conquest. 
They withdrew from the League of Nations. They began beefing up the 
Army. They denounced the naval disarmament treaties and began an 
almost frantic ship-building race . Nor did they neglect, as part of their 
war-making capital, their cryptographic assets . In 1934, their Navy 
purchased a commercial German cipher machine called the Enigma ; that 
same year, the Foreign Office adopted it, and it evolved into the most 
secret Japanese system of cryptography. A variety of other cryptosystems 
supplemented it. The War, Navy, and Foreign ministries shared the 
superenciphered numerical HATO code for intercommunication. Each 
ministry also had its own hierarchy of codes. The Foreign Office, for 
example, employed four main systems, each for a specific level of 
security, as well as some additional miscellaneous ones . 
Meanwhile , the modern-style shoguns speared into defenseless China , 
sank the American gunboat Panay, raped Nanking, molested American 
hospitals and missions in China, and raged at American embargoes on 
oil and steel scrap. It became increasingly evident that Nippon's march of 
aggression would eventually collide with American rectitude. The 
mounting curve of tension was matched by the rising output of the 
American cryptanalytic agencies. A trickle of MAGIC in 1936 had become a 
stream in 1940. Credit for this belongs largely to Major General Joseph 
O. Mauborgne, who became Chief Signal Officer in October, 1937. 
Mauborgne had long been interested in cryptology. In 1914, as a 
young first lieutenant, he achieved the first recorded solution of a cipher 
known as the Playfair, then used by the British as their field cipher. He 
described his technique in a 19-page pamphlet that was the first 
publication on cryptology issued by the United States government. In 
World War I, he put together several cryptographic elements to create the 
only theoretically unbreakable cipher, and promoted the first automatic  
cipher machine, with which the unbreakable cipher was associated. 
When he became head of the Signal Corps, he immediately set about 
augmenting the important cryptanalytic activities. He established the 
S.I.S. as an independent division reporting directly to him, enlarged its 
functions, set up branches, started correspondence courses, added 
intercept facilities , increased its budget , and put on more men. In 1939, 
when war broke out in Europe , S.I.S. was the first agency in the War 
Department to receive more funds, personnel, and space. Perhaps most 
important of all, Mauborgne's intense interest inspired his men to 
outstanding accomplishments. More and more codes were broken , and 
as the international situation stimulated an increasing flow of intercepts, 
the MAGIC intelligence approached flood stage. 
Mauborgne retired in September, 1941, leaving an expanded  
organization running with smooth efficiency. By then the Japanese had 
completed the basic outline for a dawn attack on Pearl Harbor. The plan 
had been conceived in the fertile brain of Admiral Isoroku Yamamoto, 
Commander-in-Chief Combined Fleet , Imperial Japanese Navy. Early in 
the year, he had ordered a study of the operation, contending that "If we 
have war with the United States, we will have no hope of winning unless  
the United States fleet in Hawaiian waters can be destroyed." By May 
1941, studies had shown the feasibility of a surprise air attack, statistics 
had been gathered, and operational planning was under way. 
In the middle of that month , the U.S. Navy took an important step in 
the radio intelligence field. It detached a 43-year-old lieutenant 
commander from his intelligence berth aboard U.S.S. Indianapolis and 
assigned him to reorganize and strengthen the radio intelligence unit at 
Pearl Harbor. The officer was Joseph John Rochefort, the only man in the 
Navy with expertise in three closely related and urgently needed fields: 
cryptanalysis, radio, and the Japanese language. Rochefort, who had 
begun his career as an enlisted man, had headed the Navy's 
cryptographic section from 1925 to 1927. Two years later, a married man 
with a child , he was sent, because of his outstanding abilities, as a 
language student to Japan, a hard post to which ordinarily only bachelor 
officers were sent. This three-year tour was followed by half a year in 
naval intelligence; most of the next eight years were spent at sea. 
Finally , in June of 1941, Rochefort took over the command of what 
was then known as the Radio Unit of the 14th Naval District in Hawaii. 
To disguise its functions he renamed it the Combat Intelligence Unit. His 
mission was to find out, through communications intelligence, as much 
as possible about the dispositions and operations of the Japanese Navy. 
To this end he was to cryptanalyze all minor and one of the two major 
Japanese naval crypto -systems. 
His chief target was the flag officers' system, the Japanese Navy's 
most difficult and the one in which it encased its most secret 
information. From about 1926 to the end of November, 1940, previous 
editions had provided the U.S. Navy with much of its information on the 
Japanese Navy. But the new version—a four- character code with a 
transposition superencipherment—was stoutly resisting the best efforts 
of the Navy's most skilled cryptanalysts, and Rochefort was urged to 
concentrate on it. The other major system, the main fleet cryptographic 
system, the most widely used, comprised a code with five digit code-
numbers to which were added a key of other numbers to complicate the 
system. The Navy called it the "five numeral system," or, more formally, 
JN25b—the JN for "Japanese Navy," the 25 an identifying number, the b 
for the second (and current ) edition. Navy cryptanalytic units in 
Washington and the Philippines were working on this code. Rochefort's 
unit did not attack this but did attack the eight or ten lesser codes 
dealing with personnel, engineering , administration, weather , fleet 
exercises. 
But cryptanalysis was only part of the unit's task . The great  majority  
of its 100 officers and men worked on two other aspects of radio 
intelligence—direction-finding and traffic analysis. 
Direction-finding locates radio transmitters. Since radio signals are 
heard best when the receiver points at the transmitter, sensitive  
antennas can find the direction from which a signal is coming by 
swinging until they hear it at its loudest. If two direction- finders take 
bearings like that on a signal and a control center draws the lines of 
direction on a map, the point at which they cross marks the position of 
the transmitter. Such a fix can tell quite precisely where, for example, a 
ship is operating . Successive fixes can plot its course and speed . 
To exploit this source of information, the Navy in 1937 established the 
Mid-Pacific Strategic Direction-Finder Net. By 1941, high- frequency  
direction-finders curved in a gigantic arc from Cavite in the Philippines 
through Guam , Samoa, Midway, and Hawaii to Dutch Harbor, Alaska. 
The 60 or 70 officers and men who staffed these outposts reported their 
bearings to Hawaii, where Rochefort's unit translated them into fixes. For 
example, on October 16, the ship with call-sign KUNA 1 was located at 
10.7 degrees north latitude, 166.7 degrees east longitude—or within 
Japan's mandated islands . 
These findings did not serve merely to keep an eye on the day-to-day 
locations of Japanese warships. They also formed the basis of the even 
more fruitful technique of traffic analysis. Traffic analysis deduces the 
lines of command of military or naval forces by ascertaining which radios 
talk to which. And since military operations are usually accompanied by 
an increase in communications, traffic analysis can infer the imminence 
of such operations by watching the volume of traffic. When combined 
with direction-finding, it can often approximate the where and when of a 
planned movement . 
Radio intelligence thus maintains a long-range, invisible, and 
continuous surveillance of fleet movements and organization, providing a 
wealth of information at a low cost . Of course it has its limitations. A 
change of the call-signs of radio transmitters can hinder it. The sending 
of fictitious messages can befuddle it. Radio silences can deafen it. But it 
cannot be wholly prevented except by unacceptable restrictions on 
communications. Hence the Navy relied increasingly on it for its 
information on Japanese naval activities as security tightened in Japan 
during 1941, and almost exclusively after July , when the President's 
trade-freezing order deprived the Navy of all visual observations of 
Japanese ships not on the China coast. 
It was in July that a Japanese tactic set up a radio pattern that was 
later to deceive the Combat Intelligence Unit. The Nipponese militarists 
had decided to take advantage of France 's defeat and occupy French  
Indochina. The Naval preparations for the successful grab were clearly  
indicated in the radio traffic, which went through the usual three stages 
that preceded major Japanese operations. First appeared a heavy flurry 
of messages. The Commander-in-Chief Combined Fleet busily originated 
traffic, talking with many commands to the south , thereby indicating the 
probable direction of his advance. Then came a realignment of forces. In 
the lingo of the tranalysis people, certain chickens (fleet units) no longer 
had their old mothers (fleet commanders). Call-sign NOTA 4, which 
usually communicated with OYO 8, now talked mostly with ORU 6. 
Accompanying this was a considerable confusion in the routing of 
messages, with frequent retransmissions caused by the regrouping: 
Admiral z not here; try Second Fleet. Then followed the third phase: radio 
silence. The task force was now under way. Messages would be 
addressed to it, but none would emanate from it. 
During all this, however, not only were no messages heard from the 
aircraft carriers, none were sent to them, either. This blank condition  
exceeded radio silence, which suppresses traffic in only one direction—
from the mobile force—not in both. American intelligence reasoned that 
the carriers were standing by in home waters as a covering force in case 
of counterattack, and that communications both to and from them were 
not heard because they were being sent out by short-range, low-powered 
transmissions that died away before reaching American receivers. Such a 
blank condition had obtained in a similar tactical situation in February. 
American intelligence had drawn the same conclusions then and had 
been proven right. Events soon confirmed the July assessment as well. 
Twice, then, a complete blank of carrier communications combined with 
indications of a strong southward thrust had meant the presence of the 
carriers in Empire waters. But what happened in February and July was 
not necessarily what would happen in December. 
 
During the summer and fall of 1941, the pressure of events molded 
America's two cryptanalytic agencies closer and closer to the form they 
were to have on December 7. The Signal Intelligence Service, which had 
181 officers, enlisted men, and civilians in Washington and 150 at 
intercept stations in the field on Pearl Harbor Day, had been headed 
since March by Lieutenant Colonel Rex W. Minckler, a career Signal 
Corps officer. Friedman served as his chief technical assistant. S.I.S. 
comprised the Signal Intelligence School, which trained Regular Army 
and Reserve officers in cryptology, the 2nd Signal Service Company, 
which staffed the intercept posts, and four Washington sections of the 
S.I.S. proper: the A, or administrative, which also operated the tabulating 
machinery; the B, or cryptanalytic; the c, or cryptographic, which 
prepared new U.S. Army systems, studied the current systems for 
security, and monitored Army traffic for security violations; and the D, or 
laboratory , which concocted secret inks and tested suspected 
documents . 
The B section, under Major Harold S. Doud, a West Point graduate , 
had as its mission the solution of the military and diplomatic systems 
not only of Japan but of other countries. In this it apparently achieved at 
least a fair success , though no Japanese military systems—the chief of 
which was a code employing four-digit codenum-bers—were readable by 
December 7 because of a paucity of material. Doud's technical assistant 
was a civilian, Frank B. Rowlett, one of the three original junior 
cryptanalysts hired in 1930. The military man in charge of Japanese 
diplomatic solutions was Major Eric Svensson. 
The Navy's official designation of OP-20-G indicated that the agency 
was the G section of the 20th division of OPNAV, the Office of the Chief of 
Naval Operations, the Navy's headquarters establishment . The 20th 
division was the Office of Naval Communications, and the G section was 
the Communication Security Section. This carefully chosen name 
masked its cryptanalytic activities, though its duties did include U. S. 
Navy cryptography. 
Its chief was Commander Laurence F. Safford, 48, a tall , blond  
Annapolis graduate who was the Navy's chief expert in cryptology. In 
January , 1924, he had become the officer in charge of the newly created 
research desk in the Navy's Code and Signal Section. Here he founded  
the Navy's communication-intelligence organization. After sea duty from 
1926 to 1929, he returned to cryptologic activities for three more years, 
when sea duty was again made necessary by the "Manchu" laws , which 
required officers of the Army and Navy to serve in the field or at sea to 
win promotion . He took command of OP-20-G in 1936. One of his 
principal accomplishments before the outbreak of war was the 
establishment of the Mid-Pacific Strategic Direction-Finder Net and of a 
similar net for the Atlantic , where it was to play a role of immense 
importance in the Battle of the Atlantic against the U-boats. 
Safford's organization enjoyed broad cryptologic functions. It printed 
new editions of codes and ciphers and distributed them, and contracted 
with manufacturers for cipher machines. It developed new systems for 
the Navy. It comprehended such subsections as GI, which wrote reports 
based on radio intelligence from the field units, and GL, a record-keeping 
and historical-research group. But its main interest centered on 
cryptanalysis. 
This activity was distributed among units in Washington, Hawaii, and 
the Philippines. Only Washington attacked foreign diplomatic systems 
and naval codes used in the Atlantic theater (primarily German). 
Rochefort had primary responsibility for the Japanese naval systems. 
The Philippines chipped away at JN25 and did some diplomatic 
deciphering, with keys provided by Washington. That unit, which like 
Rochefort's was attached for administrative purposes to the local naval 
district (the 16th ), was installed in a tunnel of the island fortress of 
Corregidor. It was equipped with 26 radio receivers, apparatus for 
intercepting both high- and low-speed transmissions, a directionfinder, 
and tabulating machinery. Lieutenant Rudolph J. Fabian , 33, an 
Annapolis graduate who had had three years of radio intelligence 
experience in Washington and the Philippines, commanded. The 7 
officers and 19 men in his cryptanalytic group exchanged possible 
recoveries of JN25b codegroups with Washington and with a British 
group in Singapore; each group also had a liaison man with the other. 
Of the Navy's total radio-intelligence establishment of about 700 
officers and men, two thirds were engaged in intercept or direction-
finding activities and one third— including most of the 80 officers—in 
cryptanalysis and translation. Safford sized up the personnel of his three 
units this way: Pearl Harbor had some of the best officers, most of whom  
had four or five years of radio intelligence experience; the crew at 
Corregidor, which in general had only two or three years' experience, was 
"young, enthusiastic, and capable"; Washington—responsible for both 
overall supervision and training—had some of the most experienced  
personnel, with more than ten years' experience, and many of the least: 
90 per cent of the unit had less than a year's experience. 
Under Safford in the three subsections most closely involved with 
cryptanalysis were Lieutenant Commanders George W. Welker of GX, the 
intercept and direction-finding subsection, Lee W. Parke of GY, the 
cryptanalytical subsection, and Kramer of GZ, the translation and 
dissemination subsection. GY attacked new systems and recovered new 
keys for solved systems, such as PURPLE. But while it made the initial 
breaks in code solutions, the detailed recovery of codegroups (which was 
primarily a linguistic problem as compared to the more mathematical 
cipher solutions) was left to GZ. Four officers in GY,  assisted by chief petty  
officers, stood round -the-clock watches. Senior watch officer was 
Lieutenant (j.g.) George W. Lynn ; the others were Lieutenants (j.g.) 
Brotherhood, Pering, and Allan A. Murray . GY had others on its staff , 
such as girl typists who also did the simple deciphering of some 
diplomatic messages after the watch officers and other cryptanalysts had 
found the keys. 
Kramer was in an odd position. Though he worked in OP-20-GZ, he 
was formally attached to OP-16-F2—the Far Eastern Section of the Office 
of Naval Intelligence. This arrangement was intended in part to throw off 
the Japanese, who might have inferred some measure of success in 
codebreaking if a Japanese-language officer like Kramer were assigned to 
communications, in part to have an officer with a broad intelligence 
background distribute MAGIC so that he could answer the recipients' 
questions. Kramer, 38, who had studied in Japan from 1931 to 1934, 
had had two tours in O.N.I, proper before being assigned full time to GZ 
in June, 1940. An Annapolis graduate, chess fan, and rifle marksman, 
he lived in a world in which everything had one right way to be done. He 
chose his words with almost finicky exactness (one of his favorites was 
"precise"); he kept his pencil mustache trimmed to a hair ; he filed his 
papers tidily; he often studied his MAGIC intercepts several times over 
before delivering them. Included in this philosophy was his duty. He 
performed it with great responsibility, intelligence, and dedication . 
 
The first task of OP-20-G and of S.I.S. was to obtain intercepts. And in 
peacetime America that was not easy . 
Section 605 of the Federal Communications Act of 1934, which 
prohibits wiretaps, also prohibits the interception of messages between 
foreign countries and the United States and territories. General Malin 
Craig , Chief of Staff from 1937 to 1939, was acutely aware of this, and 
his attitude dampened efforts to intercept the Japanese diplomatic 
messages coming into America. But after General George C. Marshall 
succeeded to Craig's post, the exigencies of national defense relegated 
that problem in his mind to the status of a legalistic quibble. The crypt-
analytic agencies pressed ahead in their intercept programs. The extreme  
secrecy in which they were cloaked helped them avoid detection. They 
concentrated on radio messages, since the cable companies , fully 
cognizant of the legal restrictions, in general refused to turn over any 
foreign communications to them. Consequently, 95 per cent of the 
intercepts were radio messages. The remainder was split between cable 
intercepts and photographs of messages on file at a few cooperative cable 
offices. 
To pluck the messages from the airwaves, the Navy relied mainly on 
its listening posts at Bainbridge Island in Puget Sound; Winter Harbor, 
Maine; Cheltenham, Maryland; Heeia, Oahu ; and Corregidor and to a 
lesser degree on stations at Guam; Imperial Beach , California ; 
Amagansett, Long Island and Jupiter , Florida . Each station was assigned 
certain frequencies to cover. Bainbridge Island, which was called Station 
S, copied solid the schedule of Japanese government messages between 
Tokyo and San Francisco . Its two sound recorders guarded the 
radiotelephone band of that circuit; presumably it was equipped to 
unscramble the relatively simple sound inversion that then provided 
privacy from casual eavesdropping. Diplomatic messages were 
transmitted almost exclusively by commercial radio using roman letters. 
The naval radiograms, however, employed the special Morse code devised 
for kata kana , a syllabic script of Japanese. The Navy picked these up 
with operators trained in Japanese Morse and recorded them on a 
special typewriter that it had developed for the roman-letter equivalents 
of the kana characters . The Army's stations, called Monitor Posts, were: 
No. 1, Fort Hancock , New Jersey ; No. 2, San Francisco; No. 3, Fort Sam 
Houston, San Antonio ; No. 4, Panama; No. 5, Fort Shafter, Honolulu; No. 
6, Fort Mills , Manila ; No. 7, Fort Hunt, Virginia ; No. 9, Rio de Janeiro . 
At first both services airmailed messages from their intercept posts to 
Washington. But this proved too slow . The Pan-American Clipper, which 
carried Army intercepts from Hawaii to the mainland, departed only once 
a week on the average , and weather sometimes caused cancellations, 
forcing messages to be sent by ship. As late as the week before Pearl 
Harbor, two Army intercepts from Rio did not reach Washington for 
eleven days . Such delays compelled the Navy to install teletypewriter 
service in 1941 between Washington and its intercept stations in the 
continental U.S. The station would perforate a batch of intercepts onto a 
teleptype tape, connect with Washington through a teletypewriter 
exchange, and run the tape through mechanically at 60 words per 
minute, cutting toll charges to one third the cost of manually sending 
each message individually. Outlying stations of both the Army and Navy 
picked out Japanese messages bearing certain indicators, enciphered the 
Japanese cryptograms in an American system, and radioed them to 
Washington. The reencipherment was to keep the Japanese from 
knowing of the extensive American cryptanalytic effort. Only the three 
top Japanese systems were involved in this expensive radio 
retransmission: PURPLE, RED (a machine system that antedated PURPLE, 
which had supplanted it at major embassies, but that was still in use for 
legations such as Vladivostok), and the J series of enciphered codes. The 
Army did not install a teletype for intercepts from its continental posts 
until the afternoon of December 6, 1941; the first messages (from San 
Francisco) were received in the early morning hours of December 7. 
The intercept services missed little. Of the 227 messages pertaining to 
Japanese-American negotiations sent between Tokyo and Washington 
from March to December, 1941, all but four were picked up. 
In Honolulu, where a large Japanese population produced nightmares 
of antlike espionage and potential sabotage, the 14th Naval District's 
intelligence officer, Captain Irving S. Mayfield, had long sought to obtain 
copies of the cablegrams of Consul General Nagao Kita . If Rochefort's 
unit could solve these, Mayfield figured, he might know better which 
Japanese to shadow and what information they sought. 
His intuitions were sound. On March 27, 1941, not two weeks after 
Mayfield himself took up his duties, a young ensign of the Imperial 
Japanese Navy, 25-year-old Takeo Yoshikawa, who had steeped himself 
in information about the American Navy, arrived in Honolulu to serve as 
Japan's only military espionage agent covering Pearl Harbor. Under the 
cover-name "Tadasi Morimura," he was assigned to the consulate as a 
secretary. He promptly made himself obnoxious—and drew suspicion 
upon himself within the consulate staff—by coming to work late or not at 
all, getting drunk frequently, having women in his quarters overnight, 
and even insulting the consul himself on occasion. But he managed to 
tour the islands, and within a month was sending such messages as: 
"Warships observed at anchor on the llth [of May, 1941] in Pearl Harbor 
were as follows: Battleships , 11: Colorado, West Virginia, California, 
Tennessee . . . ."  These were sent in the consulate's diplomatic systems, 
not in naval code. 
But Mayfield's hopes of peering into these secret activities through the 
window of a broken code were stymied by the refusal of the cable offices 
to violate the statute against interception. So when David Sarnoff, 
president of the Radio Corporation of America, vacationed in Hawaii, 
Mayfield spoke to him. It was subsequently arranged that thenceforth 
R.C.A.'s Japanese consulate messages would be quietly given to the 
naval authorities. But the consulate rotated its business among the 
several cable companies in Honolulu, and R.C.A.'s turn was not due until 
December 1. 
 
In Washington, however, intercepts overwhelmed GY and S.I.S. The 
tiny staff of cryptanalysts simply could not cope with all of them 
expeditiously. This difficulty was resolved in two ways. 
One was to cut out duplication of effort. At first, both services solved 
all their Japanese diplomatic intercepts. But beginning more than a year 
before Pearl Harbor, messages originating in Tokyo on odd-numbered 
days of the month were handled by the Navy, those on even days, by the 
Army. Each began breaking the messages sent in from its own intercept 
stations until it reached the Tokyo date of origin; it would then retain 
them or send them over as the dates indicated. The cryptanalysts utilized 
the extra time to attack as-yet-unbroken systems and to clean up 
backlogs. 
The other method was to concentrate on the important intercepts and 
let the others slide, at least until the important ones were completed. But 
how can a cryptanalyst tell which messages are important until he has 
solved them? He cannot, but he can assume that messages sent in the 
more secret systems are the more important. All dispatches cannot be 
transmitted in a single system because the huge volume of traffic would 
enable cryptanalysts to break it too quickly. Hence most nations set up a 
hierarchy of systems, reserving the top ones for their vital needs. 
Japan was no exception. Though her Foreign Office employed an 
almost bewildering variety of different codes, resorting, from time to time, 
to the Yokohama Specie Bank 's private code, a Chinese ideographic code 
list, and codes bearing kata kana names, such as TA, JI, or HEN, it relied 
in the main on four systems. American cryptanalysts ranked these on 
four levels according to the inherent difficulty of their solution and the 
messages that they generally carried. Intercepts were then solved in the 
order of this priority schedule. 
Simplest of all, and hence the lowest in rank and last to be read 
(excluding plain language), was the LA code, so called from the indicator  
group LA that preceded its codetexts. LA did little more than put kata 
kana into roman letters for telegraphic transmission and to secure some 
abbreviation for cable economy. Thus the kana for ki was replaced by the 
code form CI, the kana for to by IF, the two-kana combination of ka + n by 
CE. Its two-letter codewords, all of either vowel - consonant or consonant-
vowel form and including such as ZO for 4, were supplemented by a list 
of four-letter codewords, such as TUVE for dollars, SISA for ryoji ("consul"), 
and XYGY for Yokohama. A very typical  LA message is serial 01250 from 
the Foreign Minister to Kita, dated December 4, which begins in 
translation: "The following has been authorized as the year-end bonus for 
employee typists of your office." This sort of code is generally called a 
"passport code" because it usually serves for messages covering the 
administrative routine of a mission, such as issuance of passports and 
visas. LA was a particularly simple one to solve, partly because it had 
been in effect since 1925, partly because of the regularities in its 
construction . For example, all kana that ended in e had as code 
equivalents groups beginning with A (ke = AC, se = AD), and all that began 
with k had code equivalents beginning or ending with C. Identification of 
one kana would thus suggest the identification of others. 
One rung up the cryptographic ladder was the system known to the 
Japanese as Oite and to American code-breakers as PA-K2. The PA part 
was a two- and four-letter code similar to the LA, though much more 
extensive and with codegroups disarranged. The K2 part was a 
transposition based on a keynumber. The letters from the PA encoding 
were written under this keynumber from right to left and then copied out 
in mixed order, taking first the letter under number 1, then the letter 
under number 2, until the row was completed. The process was repeated 
for successive rows. 
For example, on December 4 Yoshikawa wired the Foreign Minister 
that "At 1 o'clock on the 4th a light cruiser of the Honolulu  class hastily 
departed—Morimura." In romaji (the roman-letter version of the kata 
kana) this became 4th gogo 1 kei jun (honoruru) kata hyaku shutsu ko—
morimura. In PA, with the parentheses getting their own codegroups (OQ 
and UQ), it assumed this form: BYDH DOST JE YO IA OQ GU RA HY HY UQ VI LA YJ 
AY EC TY FI BANL, with FI indicating use four-letter code. (The code clerk  
made two errors. After encoding kata by VI, he encoded an extra ta into 
LA and an unnecessary re into TY.) This was then written under the 
keynumber from right to left, with an extra letter I as a null to complete 
the final five-letter group: 
 
10 15 11 16 2 8 1 5 17 3 7 13 19 4 18 6 12 9 14
 B  Y  D  H D  O S T  J E Y O  I A  O Q  G U  R 
 A  H  Y  H Y  U Q V  I L A Y  J A  Y E  G T  Y 
            F    I B    A N      L    I 
Transcribed line by line according to the numbers (s under 1 first, D 
under 2 second, etc.), prefixed with system indicator  GIGIG  and key 
indicator AUDOB, the message number, and the telegraphic abbreviation 
of Sikuyu ("urgent"), the message (with three more errors: the Y under 13 
became the J in CJYHH, the F under 2 became the E in IYJIE, and the T 
under 9 became the i in AUIAY) became the one actually sent over Kita's 
name: 
 
GAIMUDAIJIN TOKIO  
SIKYU 02500 GIGIG AUDOB SDEAT QYOUB DGORY HJOIQ YLAVE 
AUIAY CJYHH IYJIE ALBIN 
KITA 
 
PA-K2 did not pose much of a problem to experienced American 
cryptanalysts. ROchefort estimated that his unit could crack a PA-K2 
message in from six hours to six days, with three days a good average. 
The transposition was vulnerable because each line was shuffled 
identically; the cryptanalyst could slice a cryptogram into groups of 15 or 
17 or 19 and anagram these simultaneously until the predominant 
vowel-consonant alternation appeared on all lines; the underlying code 
could then be solved by assuming that the most frequent codegroups 
represented the most frequent kana (i, followed by ma, shi, o, etc.) and 
filling out the skeleton words that resulted. Since the system had 
remained in use for several years, this reconstruction had long been 
accomplished by the Washington agencies. Hence solution involved only 
unraveling any new transposition and, with luck , might take only a few 
hours. It could also take a few days. Primarily because of PA-K2's 
deferred position in the priority list, an average of two to four days 
elapsed between interception and translation. 
The code clerk in Honolulu enveloped Yoshikawa's final messages in 
PA-K2 only because higher -level codes had been destroyed December 2 
on orders from Tokyo. Normally, espionage reports of shipping 
movements and military activities, sent routinely by Japanese consuls 
from their posts all over the world, were framed on that next level of 
secrecy. Here prevailed a succession of codes called TSU by the Japanese 
and the J series by Americans. These were even more extensive and more 
thoroughly disarranged than PA, and they were transposed by a system of 
far greater complexity than the rather simple and vulnerable K2. 
Furthermore, the code and the transposition were changed at frequent 
intervals. Thus J17-K6 was replaced on March 1 by J18-K8, and that in 
turn by J19-K9 on August 1. 
The transposition was the real stumbling block. Like the K2, it used a 
keynumber, but it differed in being copied off vertically instead of 
horizontally, and in having a pattern of holes in the transposition blocks. 
These holes were left blank when the code groups are inscribed into the 
block. For example, letting the alphabet from A to Y serve as the code 
message: 
 
[CodeBreakers 020.jpg]
 
The letters were transcribed in columns in the order of the 
keynumbers, skipping over the blanks: BJMV EHKT NW CGORX AFILQU DPSY. 
This would be sent in the usual five-letter groups. 
The first step in solving a columnar transposition like this, but 
without blanks, is to cut the cryptogram into the approximately equal 
segments that the cryptanalyst believes represent the columns of the 
original block. The blanks vastly increase the difficulty of this essential  
first step because they vary the length of the column segments. The 
second step is to reconstruct the block by trying one segment next to the 
other until a codeword-like pattern appears . Here again the blanks, by 
introducing gaps in unknown places between the letters of the segments, 
greatly hinder the cryptanalyst. 
The problems of solving such a system are illustrated by the fact that 
J18-K8 was not broken until more than a month after its introduction . 
The cryptanalysts had to make a fresh analysis for each pattern of 
blanks and each transposition key. The key changed daily, the blank-
pattern three times a month. Hence J19-K9 solutions were frequently 
delayed. The key and pattern for November 18 were not recovered until 
December 3; those for November 28, not until December 7. On the other 
hand, solution was sometimes effected within a day or two. Success 
usually depended on the quantity of intercepts in a given key. About 10 
or 15 per cent of J19-K9 keys were never solved. 
This situation contrasts with that of PURPLE, the most secret Japanese 
system, in which all but 2 or 3 per cent of keys were recovered and in 
which most messages were solved within hours. Did the Japanese err in 
assessing the security of their systems? Yes and no. PURPLE was easier to 
keep up with once it was solved, but it was a much more difficult system 
to break in the first place than J19-K9. The solution of the PURPLE 
machine was, in fact, the greatest feat of cryptanalysis the world had yet 
known. 
 
The cipher machine that Americans knew as PURPLE bore the 
resounding official Japanese title of 97-shiki O-bun In-ji-ki. This meant 
Alphabetical Typewriter '97, the '97 an abbreviation for the year 2597 of 
the Japanese calendar , which corresponds to 1937. The Japanese 
usually referred to it simply as "the machine" or as "J,"1 the name given 
it by the Imperial Japanese Navy, which had adapted it from the German 
Enigma cipher machine and then had lent it to the Foreign Ministry, 
which, in turn, had further modified it. Its operating parts were housed 
in a drawer -sized box between two big black electrically operated 
Underwood typewriters, which were connected to it by 26 wires plugged 
into a row of sockets called a plugboard. To encipher a message, the 
cipher clerk would consult the thick YU GO book of machine keys, plug in 
the wire connections according to the key for the day, turn the four disks 
in the box so the numbers on their edges were those directed by the YU 
GO, and type out the plaintext. His machine would record that plaintext 
while the other, getting the electric impulses after the coding box had 
twisted them through devious paths, would print out the cipher-text. 
Deciphering was the same, though the machine irritatingly printed the 
plaintext in the five-letter groups of the ciphertext input. 
The Alphabetical Typewriter worked on roman letters, not kata kana. 
Hence it could encipher English as well as romaji—and also roman-letter 
codetexts, like those of the J codes. Since the machine could not 
encipher numerals or punctuation, the code clerk first transformed them 
into three-letter codewords, given in a small code list, and enciphered 
these. The receiving clerk would restore the punctuation, paragraphing, 
and so on, when typing up a finished copy of the decode. 
The coding wheels and plugboards produced a cipher of great 
difficulty. The more a cipher deviates from the simple form in which one 
ciphertext letter invariably replaces the same plaintext letter, the harder 
it is to break. A cipher might replace a given plaintext letter by five 
different ciphertext letters in rotation, for example. But the Alphabetical 
Typewriter produced a substitution series hundreds of thousands of 
letters long. Its coding wheels, stepping a space—or two, or three, or 
four—after every letter or so, did not return to their original positions to 
re-create the same series of paths, and hence the same sequence of 
substitutes, until hundreds of thousands of letters had been enciphered. 
The task of the cryptanalysts consisted primarily of reconstructing the 
wiring and switches of the coding wheels—a task made more 
burdensome by the daily change of plugboard connections. Once this 
was done, the cryptanalyst still had to determine the starting position at 
the coding wheels for each day's messages. But this was a comparatively 
simple secondary job. 
American cryptanalysts knew none of these details when the 
Japanese Foreign Office installed the Alphabetical Typewriter in its major 
embassies in the late 1930s . How, then, did they solve it? Where did they 
begin ? How did they even know that a new machine was in service, since 
the Japanese government did not announce it? 
The PURPLE machine supplanted the RED machine,2 which American 
cryptanalysts had solved, and so probably their first clue to the new 
machine was the disconcerting discovery that they could no longer read 
the important Japanese messages. At the same time, they observed new 
indicators for the PURPLE system. Clues to the system's nature came from 
such characteristics of its ciphertext as the frequency of letters, the 
percentage of blanks (letters that did not appear in a given message), and 
the nature and number of repetitions. Perhaps the codebreakers also 
assumed that the new machine comprised essentially a more 
complicated and improved version of the one it replaced. In this they 
were right. 
Their first essays at breaking into the cipher both accompanied and 
supplemented their attempts to determine the type of cipher. Their 
previous success with the RED machine and with the lesser systems had 
given them insight into the Japanese diplomatic forms of address, 
favorite phrases, and style (paragraphs were often numbered, for 
example). These provided the cryptanalysts with probable words—words 
likely to be in the plaintext— that would help in breaking the cipher. 
Opening and closing formulas , such as "I have the honor to inform Your 
Excellency" and "Re your telegram," constituted virtual cribs. Newspaper  
stories suggested the subject matter of intercepts. The State Department 
sometimes made public the full texts of diplomatic notes from Japan to 
the American government, in effect handing the cryptanalysts the 
plaintext (or its translation) of an entire dispatch. (State reportedly did 
not pass the texts of confidential notes to the cryptanalysts, though this 
would have helped them considerably and was done by other foreign 
ministries.) Japan's Foreign Office often had to circulate the same text to 
several embassies, not all of which had a PURPLE machine, and a code 
clerk might have inadvertently encoded some cables in PURPLE, some in 
other systems— which the cryptanalysts could read. A comparison of 
times of dispatch and length, and voilá!—another crib to a cryptogram. 
Errors were, as always, a fruitful source of clues. As late as November, 
1941, the Manila legation repeated a telegram "because of a mistake on 
the plugboard." How much more common must errors have been when 
the code clerks were just learning to handle the machine! The sending of 
the identical text in two different keys produces "isomorphic" 
cryptograms that yield exceedingly valuable information on the 
composition of the cipher. 
The cryptanalysts of S.I.S. and OP-20-G, then, matched these 
assumed plaintexts to their ciphertexts and looked for regularities from 
which they could derive a pattern of encipherment. This kind of work, 
particularly in the early stages of a difficult cryptanalysis, is perhaps the 
most excruciating, exasperating, agonizing mental process known to 
man. Hour after hour, day after day, sometimes month after month, the 
cryptanalyst tortures his brain to find some relationship between the 
letters that hangs together, does not dead-end in self-contradiction, and 
leads to additional valid results . 
The codebreakers attacking the new Japanese mechanism went just 
so far—and for months could not push on further. As William Friedman 
recalled, "When the PURPLE system was first introduced it presented an 
extremely difficult problem on which the Chief Signal Officer [Mauborgne] 
asked us to direct our best efforts. After work by my associates when we 
were making very slow progress, the Chief Signal Officer asked me 
personally to take a hand. I had been engaged largely in administrative 
duties up to that time, so at his request I dropped everything else that I 
could and began to work with the group." 
Lighting his way with some of the methods that he himself had 
developed, he led the cryptanalysts through the murky PURPLE 
shadowland. He assigned teams to test various hypotheses. Some 
prospected fruitlessly, their only result a demonstration that success lay 
in another direction. Others found bits and pieces that seemed to make 
sense. (OP-20-G cooperated in this work, with Harry L. Clark making 
especially valuable contributions, but S.I.S. did most of it.) Friedman and 
the other codebreakers began to segregate the ciphertext letters into 
cycles representing the rotation of the coding wheels—gingerly at first, 
then faster and faster as the evidence accumulated. The polyalphabetic 
class of ciphers, to which PURPLE belonged, is based ultimately upon an 
alphabet table, usually 26 letters by 26. To reconstruct the PURPLE tables, 
the cryptanalysts employed both direct and indirect symmetry of 
position— names only slightly less forbidden than the methods they 
denote. Errors, caused perhaps by garbled interceptions or simple 
mistakes in the cryptanalysis, jarred these delicate analyses and delayed 
the work. But slowly it progressed. A cryptanalyst, brooding sphinxlike 
over the cross-ruled paper on his desk, would glimpse the skeleton of a 
pattern in a few scattered letters; he tried fitting a fragment from another 
recovery into it; he tested the new values that resulted and found that 
they produced acceptable plaintext; he incorporated his essay into the 
over-all solution and pressed on. Experts in Japanese filled in missing 
letters; mathematicians tied in one cycle with another and both to the 
tables. Every weapon of cryptanalytic science—which in the stratospheric 
realm of this solution drew heavily upon mathematics, using group 
theory, congruences, Poisson distributions—was thrown into the fray. 
Eventually the solution reached the point where the cryptanalysts had 
a pretty good pencil-and-paper analog of the PURPLE machine. S.I.S. then 
constructed a mechanism that would do automatically what the 
cryptanalysts could do manually with their tables and cycles. They 
assembled it out of ordinary hardware and easily available pieces of 
communication equipment , such as the selector switches used for 
telephones. It was hardly a beautiful piece of machinery, and when not 
running just right it spewed sparks and made loud whirring noises. 
Though the Americans never saw the 97-shiki O-bun In-ji-ki, their 
contraption bore a surprising physical resemblance to it, and of course 
exactly duplicated it cryptographically. 
S.I.S. handed in its first complete PURPLE solution in August of 1940, 
after 18 or 20 months of the most intensive analysis. In looking back on 
the effort that culminated in this, the outstanding cryptanalytic success 
in the whole history of secret writing up to its time, Friedman would say 
generously: 
 
Naturally this was a collaborative, cooperative effort on the part 
of all the people concerned. No one person is responsible for the 
solution, nor is there any single person to whom the major share of 
credit should go. As I say, it was a team, and it was only by very 
closely coordinated teamwork that we were able to solve it, which 
we did. It represents an achievement of the Army cryptanalytic 
bureau that, so far as I know, has not been duplicated elsewhere, 
because we definitely know that the British cryptanalytic service 
and the German cryptanalytic service were baffled in their 
attempts and they never did solve it. 
 
Friedman, was, despite his partial disclaimer, the captain of that 
team. The solution had taken a terrific toll. The restless turning of the 
mind tormented by a puzzle , the preoccupation at meals, the insomnia , 
the sudden wakening at midnight, the pressure to succeed because 
failure could have national consequences , the despair of the long weeks 
when the problem seemed insoluble, the repeated dashings of uplifted 
hopes, the mental shocks, the tension and the frustration and the 
urgency and the secrecy all converged and hammered furiously upon his 
skull . He collapsed in December. After three and a half months in Walter 
Reed General Hospital recovering from the nervous breakdown, he 
returned to S.I.S. on shortened hours, working at first in the more 
relaxed area of cryptosecurity. By the time of Pearl Harbor he was again 
able to do some cryptanalysis, this time of German systems. 
OP-20-G contributed importantly to the ease and speed of daily PURPLE 
solutions when 27-year-old Lieutenant (j.g.) Francis A. Raven discovered  
the key to the keys. After a number of PURPLE messages has been solved, 
Raven observed that the daily keys within each of the three ten-day 
periods of a month appeared to be related. He soon found that the 
Japanese simply shuffled the first day's key to form the keys for the next 
nine days, and that the nine shuffling patterns were the same in all the 
ten-day periods. Raven's discovery enabled the cryptanalysts to predict  
the keys for nine out of ten days. The cryptanalysts still had to solve for 
the first day's key by straightforward analysis, but this task and its 
delays were eliminated for the rest of the period. Furthermore, knowledge  
of the shuffles enabled the codebreakers to read all the traffic of a period 
even though they could solve only one of the daily keys. 
This fine piece of work, on the shoulders of the tremendous initial 
Friedman-S.I.S. effort, resulted in the paradoxical situation of Americans 
reading the most difficult Japanese diplomatic system more quickly and 
easily than some lower- grade systems. They also became very facile in 
reading two-step systems in which PURPLE superenciphered an already  
coded message. The Japanese did this from time to time to provide extra 
security, usually with the CA code, the personal code of an ambassador or 
head of mission. A year after S.I.S. handed in its first PURPLE solution, the 
cryptanalysts solved a message enciphered in "the highest type of secret 
classification used by the Japanese Foreign Office." The message was 
first enciphered in CA; this was then juggled according to the K9 
transposition (normally used with the J19 code), and the transposed 
codetext was then enciphered on the PURPLE machine. The solution, 
which on the basis of the number of combinations involved might have 
been expected to take geologic eons, was completed in just four days. 
 
The intercepts ordinarily needed to be translated, and translation was 
the bottleneck of the MAGIC production line. Interpreters of Japanese were 
even scarcer than expert cryptanalysts. Security precluded employing 
Nisei or any but the most trustworthy Americans. Through prodigious 
efforts in 1941 the Navy doubled its GZ translation staff —to six. These 
included three whom Kramer called "the most highly skilled Occidentals 
in the Japanese language in the world." 
But ability in standard Japanese alone did not suffice. Each 
translator had to have at least a year's experience in telegraphic 
Japanese as well before he could be trusted to come through with the 
correct interpretation of a dispatch. This is because telegraphic Japanese 
is virtually a language within a language, and, as McCollum, himself a 
Japanese-language officer, explained, "the so-called translator in this 
type of stuff almost has to be a cryptographer himself. You understand  
that these things come out in the form of syllables, and it is how you 
group your syllables that you make your words. There is no punctuation. 
"Now, without the Chinese ideograph to read from, it is most difficult 
to group these things together. That is, any two sounds grouped together 
to make a word may mean a variety of things. For instance , 'ba' may 
mean horses or fields, old women, or my hand, all depending on the 
ideographs with which it is written. On the so-called translator is forced 
the job of taking unrelated syllables and grouping them into what looks  
to him to be intelligible words, substituting then such of the Chinese 
ideographs necessary to pin it down, and then going ahead with the 
translation, which is a much more difficult job than simple translation." 
Hence the situation of Mrs. Dorothy Edgers. She had lived for thirty  
years in Japan and had a diploma from a Japanese school to teach  
Japanese to Japanese students up to high-school level. Yet, because she 
had only two weeks' experience in GZ at the time of Pearl Harbor, Kramer 
considered her "not a reliable translator" in this field. And on the 
important messages, only reliable translators could be used. To unclog 
this bottleneck, messages in the minor systems were given only a partial 
translation. If a translator saw that they dealt with administrative trivia, 
they were frequently not formally translated at all. 
With manifold streamlinings like that, with enlarged staffs, with the 
fluidity gained by experience, OP-20-G and S.I.S. gradually increased the 
speed and quality of their output. In 1939, the agencies had often 
required three weeks to funnel a message from interceptor to recipient . 
In the latter part of 1941 the process sometimes took as little as four 
hours. Occasionally an agency broke down a late intercept that bore on a 
point of Japanese-American negotiations and rushed it to the Secretary 
of State an hour before he was to meet with the Japanese ambassadors. 
Volume attained overwhelming proportions. By the fall of 1941, 50 to 75 
messages a day sluiced out of the two agencies, and at least once the 
quantity swelled to 130. Some of these messages ran to 15 typewritten 
pages. 
The top-echelon recipients of MAGIC clearly could not afford the time to 
read all this traffic. Much of it was of secondary importance anyway. 
Kramer and Colonel Rufus S. Bratton, army G-2 Far Eastern Section 
chief, winnowed the wheat from this chaff. Reading the entire output, 
they chose an average of 25 messages a day for distribution. At first 
Kramer supplemented his translations with gists for recipients too busy  
to read every word of the actual intercepts, starring the important ones, 
but he abandoned these in mid-November under the pressure of getting 
out the basic material. Bratton, who had been delivering summaries of 
MAGIC in the form of Intelligence Bulletins, began on August 5 to 
distribute MAGIC verbatim at Marshall's orders. This, however, had the 
effect of increasing the volume. Marshall complained that to absorb every 
word of it he would have had to "retire as Chief of Staff and read every 
day." To save the recipients' time, Bratton checked the important 
messages on a list in the folder with a red pencil; Kramer slid paper clips 
onto them. The recipients always read the flagged messages; the others 
they did not always study, but they did skim them. 
Distribution was usually made twice a day. Intercepts that had come 
in overnight went out in the morning, those processed during the day 
went out at the end of the afternoon. Especially important messages were 
delivered at once, often to the recipients' homes if late in the evening . 
Each agency sent its MAGIC copies on to the other with exemplary 
promptitude, despite a natural competition between them. 
As Bratton put it: "I was further urged on by the fact that if the Chief 
of Naval Operations ever got one of these things before General Marshall 
did and called him up to discuss it on the telephone with him, and the 
General hadn't gotten his copy, we all caught hell." (Marshall demurred: 
"I don't think I gave anybody hell much.") 
Delivery to the White House and the State Department incurred  
difficulties. Under an agreement of January 23, the Army and Navy at 
first alternated in servicing the two. The Army, however, discontinued its 
deliveries to the White House after its turn in May, partly because a 
military aide made a security bungle, partly because it felt that these 
diplomatic matters should go to the President through the State 
Department. The Navy continued its deliveries through the President's 
naval aide, Captain John R. Beardall, though once in the summer 
Kramer himself carried a particularly "hot" message—probably dealing 
with negotiations the next day—to Roosevelt . Near the end of September, 
a month originally scheduled for Army delivery, during which nothing 
was delivered to the White House, the President said he wanted to see 
the intercept information. In October naval intelligence sent him 
memoranda based on MAGIC, but on Friday , November 7, Roosevelt said 
he wanted to see MAGIC itself. Beardall told him that it was an Army 
month. The President replied that he knew that and that he was either 
seeing MAGIC or getting information on it from Hull, but that he still 
wanted to see the original intercepts. He feared that condensing them 
would distort their meaning . On Monday , a conference agreed that the 
Navy would furnish the White House with MAGIC and the Army the State 
Department. At 4:15 p.m., Wednesday , November 12, Kramer made the 
first distribution to the White House under this system. 
Thus, by the fall of 1941, MAGIC was being demanded at the topmost 
level of government. It had become a regular and vital factor in the 
formation of American policy . Hull, who looked upon MAGIC "as I would a 
witness who is giving evidence against his own side of the case," was "at 
all times intensely interested in the contents of the intercepts." The chief 
of Army intelligence regarded MAGIC as the most reliable and authentic 
information that the War Department was receiving on Japanese 
intentions and activities. The Navy war plans chief thought that MAGIC, 
which was largely diplomatic at this time, affected his estimates by about 
15 per cent. The high officials not only read MAGIC avidly and discussed it 
at their conferences, they acted upon it. Thus the decision to set up the 
command of United States Army Forces, Far East, which was headed by 
General MacArthur , stemmed in part from intercepts early in 1941 
showing that Germany was pressuring Japan to attack Britain in Asia in 
the hope of involving the United State in the war; on the basis of this 
information, the command was created in July to deter Japan by 
enhancing American prestige in the Western Pacific—and it is a fact that 
Japan did not then comply with Germany's wishes. 
The intricate mechanism of the American cryptanalytic effort pumped 
MAGIC to its eager recipients smoothly, speedily, and lavishly. Messages 
flew back and forth along the monitor channels as if along nerve cells. 
Intercepts poured into Washington with less and less of a time lag. S.I.S. 
and GY grew increasingly adept at solution; the translators picked out the 
important messages ever more surely. Bratton and Kramer hustled from 
place to place with their locked briefcases. MAGIC gushed forth in 
profusion. So effectively did the cryptanalytic agencies perform that 
Marshall could say of this "priceless asset ," this most complete and up-
to-the-minute intelligence that any nation had ever had concerning a 
probable enemy, this necromantic gift of the gods of which one could 
apparently never have enough, that "There was too much of it." 
 
2.  One Day of Magic:  II 
 
IN October the cabinet of Prince Konoye fell, and the Emperor summoned 
General Hideki Tojo to form a new government. One of the first acts of 
the new Foreign Minister, Shigenori Togo , was to call in the chief of the 
cable section. Togo, remembering a book that Herbert O. Yardley had 
written disclosing his 1920 solution of Japanese diplomatic codes, asked 
the cable chief, Kazuji Kameyama, whether their current diplomatic 
communications were secure. Kameyama reassured him. "This time," he 
said, "it's all right." 
With the assumption of total power by the militarists under Tojo, the 
last real hopes for peace died. Almost at once, events began to slide 
toward war. On November 4, Tokyo sent to her ambassadors at 
Washington the text of her proposal  B, which Togo described as 
"absolutely final." The ambassadors held it while they pursued other 
avenues, even though Tokyo, on November 5, told them that "Because of 
various circumstances, it is absolutely necessary that all arrangements 
for the signing of this agreement be completed by the 25th of this 
month." 
That same day, Yamamoto promulgated Combined Fleet Top Secret 
Order Number 1, the plan for the Pearl Harbor attack. Two days later, he 
set December 8 (Tokyo time) as Y-day and named Vice Admiral Chuichi 
Nagumo as Commander, First Air Fleet—the Pearl Harbor strike force. In 
the days that followed, the 32 ships that were to compose the force 
slipped, one by one, out to sea and vanished. Far from any observation , 
they headed north to rendezvous in a bay of barren Etoforu Island, one of 
the chill , desolate Kuriles north of the four main islands of Japan. 
Behind them the ships left their regular wireless operators to carry on an 
apparently routine radio traffic in their own "fists," or sending touch , 
which is as distinctive as handwriting. 
As the force was gathering , the Foreign Office, which knew only that 
the situation was tense and was never told in advance of the time, place, 
or nature of the planned attack, prepared on open -code arrangement as 
an emergency means of notification. Tokyo sent Circular 2353 to 
Washington on November 19: 
 
Regarding the broadcast of a special message in an emergency. 
In case of emergency ( danger of cutting off our diplomatic 
relations ), and the cutting off of international communications, the 
following warning will be added in the middle of the daily Japanese 
language short- wave news broadcast: 
1)   In  case  of  Japan-U.S.   relations   in   danger: HIGASHI NO 
KAZE AME ("east wind rain") 
2)   Japan-U.S.S.R.   relations:   KITA NO KAZE  KU-MORI ("north 
wind cloudy ") 
3)  Japan-British relations:  NISHE NO KAZE HARE  ("west wind 
clear") 
This signal will be given in the middle and at the end as a 
weather forecast and each sentence will be repeated twice. When 
this is heard please destroy all code papers, etc. This is as yet to be 
a completely secret arrangement. 
Forward as urgent intelligence. 
 
This open code related the winds to the compass points in which the 
named countries stood in regard to Japan: the U.S. to the east, Russia to 
the north, England to the west. Tokyo also set up an almost similar code 
for use in the general intelligence (not news) broadcasts. 
As the secret messages establishing these open codes whistled 
through the air, Navy intercept Station S at Bainbridge Island heard and 
nabbed them. The station teletyped them to GY, which identified them as 
J19 and began cryptanalysis. 
Many of the ships of the Pearl Harbor strike force had by then 
gathered in bleak Tankan Bay, where the only signs of human presence 
were a small concrete pier , a wireless shack, and three fishermen's huts. 
Snow covered the surrounding hills. In the gray twilight of November 21, 
the great carrier Zuikaku glided into the remote harbor to complete the 
roster . The force swung at anchor, awaiting the order to sortie. 
A few hours later, on November 20 (Washington time), the Japanese 
ambassador to the United States, Admiral Kichisaburo Nomura, and his 
newly arrived associate, Saburo Kurusu, presented Japan's ultimatum to 
Hull. It would have required the United States to reverse its foreign 
policy, acquiesce in further Japanese conquests, supply Japan with as 
much oil as she required for them, abandon China, and in effect 
surrender to international immorality. While Hull began drafting a reply, 
Tokyo cabled its ambassadors in message 812 that "There are reasons  
beyond your ability to guess why we wanted to settle Japanese-American 
relations by the 25th, but if within the next three or four days you can 
finish your conversations with the Americans; if the signing can be 
completed by the 29th (let me write it out for you— twenty -ninth); if the 
pertinent notes can be exchanged; if we can get an understanding with 
Great Britain and the Netherlands ; and in short if everything can be 
finished, we have decided to wait until that date. This time we mean it, 
the deadline absolutely cannot be changed. After that things are 
automatically going to happen." Two days later, Togo wirelessed: "The 
time limit set in my message No. 812 is in Tokyo time." The calendar had 
become a clock, and the clock had begun to tick. 
On November 25, Yamamoto ordered the Pearl Harbor strike force to 
sortie next day. At 6 a.m. on November 26, the 32 ships of the force—six 
carriers, two battleships, and a flock of destroyers and support vessels—
weighed anchor and sliced across the wrinkled surface of Tankan Bay. 
They steamed slightly south of east, heading into the "vacant sea"—the 
wintry North Pacific, whose wastes were undefiled by merchant tracks 
and whose empty vastness would swallow up the force. They had been 
ordered to return if detected before December 6 (Tokyo time); if 
discovered on December 7, Nagumo would decide whether or not to 
attack. Strict radio silence was enjoined. Aboard the battleship  Hiei, 
Commander Kazuyoshi Kochi, a communications officer for the force, 
removed an essential part of his transmitter and put it in a wooden box, 
which he used as a pillow . The force drove eastward through fog, gale  
winds, and high seas. No one saw them. 
 
Meanwhile, Hull, after a frantic week of drafting, consultations, and 
redraftings, had completed the American reply to Japan's proposal. It 
called upon Japan to withdraw all forces from China and Indochina and 
in return promised to unfreeze Japanese funds and resume trade. 
Nothing was said about oil. On November 26, the day that he handed it 
to Nomura, Tokyo circularized its major embassies with an open code. 
While the winds code envisioned abolition of all communication with the 
embassies, this new code—called the  INGO DENPO  ("hidden word") code —
was intended for a less critical situation. It seems to have been arranged 
at the request of the consul in Singapore in case code but not plain 
language telegrams were prohibited. It set up such equivalences as 
ARIMURA = code communications prohibited; HATTORI = relations between 
Japan and (name of  country ) are not in  accordance with  expectation ;3 
KODAMA = Japan; KUBOTA = U.S.S.R.; MINAMI =U.S.A.; and so on. "In order 
to distinguish these cables from others," Tokyo said, "the English word 
STOP will be added at the end as an indicator. (The Japanese word OWARI 
[end] will not be used.)" 
The next day, November 28, the Navy cracked the transposition for 
the J19 message of nine days earlier and learned of the winds code 
arrangement. The cryptanalytic agencies saw at once that this 
arrangement, which dispensed with the entire routine of coding, cabling, 
delivery, and decoding, could give several hours' advance warning of 
Japan's intentions. They erupted into activity to try to intercept it. This 
wrenched facilities away from the commercial (for Japanese diplomatic), 
naval, and radiotelephone circuits with which the agencies were familiar 
and put them on voice newscasts. 
The Army asked the Federal Communications Commission to listen  
for the winds code execute. Army stations at Hawaii and San Francisco 
tuned to the newscasts, as did Navy stations at Corregidor, Hawaii, and 
Bainbridge Island, and four or five along the Atlantic seaboard. Rochefort 
placed his four best language officers—Lieutenants Forrest R. Biard , J. 
R. Bromley, Allyn Cole, Jr., and G. M. Slonim—on a 24-hour watch on 
frequencies suggested by Washington and on others that his unit had 
found. The Dutch in Java and the British in Singapore listened. In 
Washington, Kramer made up some 3x5 cards for distribution to MAGIC 
recipients. They bore only the portentous phrases, "East Wind Rain: 
United States. North Wind Cloudy: Russia. West Wind Clear: England ." 
Soon plain-language intercepts were swamping GZ. Bainbridge ran up 
bills of $60 a day to send them in. Kramer and the other translators, 
already burdened, now had also to scan 100 feet of teletype paper a day 
for the execute; previously only three to five feet per week of plain 
language material had come in. The long strips were thrown into the 
wastebasket and burned after checking. Several times the GY watch 
officers telephoned Kramer at his home at night to ask him to come to 
the office and check a possible execute. It always proved false. 
Meanwhile, other signs of increasing tension were not lacking. On the 
29th, Baron Oshima in Berlin reported that the German Foreign 
Minister, Joachim von Ribbentrop, had told him, "Should Japan become 
engaged in a war against the United States, Germany, of course, would 
join the war immediately." Next day, Tokyo replied, "Say very secretly to 
them that there is extreme danger that war may suddenly break out 
between the Anglo-Saxon nations and Japan through some clash of arms 
and add that the time of the breaking out of this war may come quicker 
than anyone dreams." Both these messages were translated on December 
1, and Roosevelt considered the latter so important that he asked for a 
copy of it to keep. Kramer, after paraphrasing it for security's sake, gave 
him one. 
 
At Pearl Harbor, Rochefort had just been presented with an 
unpleasant confirmation of that tautening situation. The Japanese fleet 
reassigned its 20,000 radio call-signs at midnight, December 1—only 30 
days after the previous change. It was the first time in Rochefort's 
experience that a switch had occurred so soon after a previous one. 
The one on November 1 had been expected; it had followed by the 
usual six months the regular spring call-sign shift. With the facility born 
of long experience, Rochefort's Combat Intelligence Unit identified in 
fairly rapid order the senders and receivers of a large percentage of the 
traffic. The unit observed the rising volume and southward routing of 
messages on the 200 radio circuits of the Japanese Navy. This fitted in 
almost perfectly with the widely known Japanese buildup for what the 
world thought was a strike at Siam or Singapore. By the third week in 
November, the unit had sensed the formation of a Third Fleet task force 
and its imminent departure in the direction of those areas. Aircraft 
carriers were not addressed during this buildup, nor did they transmit. 
To Rochefort, the situation shaped up like those of February and July, 
when Japanese fleet units moved south to support the takeover in 
French Indochina while the carriers remained in home waters as a 
reserve. They were there, he felt, to protect the exposed flank of the 
Japanese forces from the American fleet, which, from its bases at Cavite 
and Pearl, could sever the supply lines of the aggressor. 
Rochefort's view was shared by fleet intelligence officer Layton. He 
knew that the two main carrier divisions had not appeared in the traffic 
for at least two weeks, and maybe three. He suspected their presence in 
home waters, but since he lacked positive indications of it, he omitted his 
presumptions from a report on the Japanese fleet that he submitted to 
Kimmel on December 1. Whereupon, Layton recalled: 
 
Admiral Kimmel said, "What! You don't know where Carrier 
Division 1 and Carrier Division 2 are!" 
I replied, "No sir, I do not. I think they are in home waters, but I 
do not know where they are. The rest of these units, I feel pretty 
confident of their location ." Then Admiral Kimmel looked at me, as 
sometimes he would, with somewhat a stern countenance and yet 
partially with a twinkle in his eye, and said: 
"Do you mean to say that they could be rounding Diamond 
Head and you wouldn't know it?" or words to that effect. My reply 
was that "I hope they would be sighted before now," or words to 
that effect. 
 
On the same day that Layton gave his report to Kimmel, the Office of 
Naval Intelligence produced a memorandum of "Japanese Fleet 
Locations" that Layton, when he saw it, considered as "dotting the i's and 
crossing the t's" of his own estimates. It placed Akagi and Kaga (Carrier 
Division 1), and Koryu and Kasuga in southern Kyushu waters, and 
Soryu and Hiryu (Carrier Division 2) and Zuikaku, Shokaku, Hosho, and 
Ryujo  at the great naval base of Kure . All this was just a more precise 
way of saying "home waters." 
These estimates were based on the November observations. The call-
sign change of December 1 obliterated the intricate communication 
networks that the radio intelligence units had so painstakingly built up 
and forced them to begin anew. The Japanese bedeviled them with new 
communication-security measures . Dispatches were sent “on the 
umbrella "—broadcast to the fleet at large and copied by all ships. This 
sort of blanket coverage made identification difficult. Multiple addresses 
were used. They sent dummy traffic, which, however, did not confuse the 
listeners. Just before the change, the communicators passed many old 
messages. Rochefort's unit spotted them, and guessed that they were 
attempts either to pad the volume or to get through to the addressee 
before the change caused routing difficulties. 
On December 2, after only two days of analyzing the new calls, 
Rochefort's unit stated in its Communications Intelligence Summary: 
"Carriers—Almost a complete blank of information of the Carriers today. 
Lack of identifications has somewhat promoted this lack of information. 
However, since over two hundred service calls have been partially 
identified since the change on the first of December and not one carrier 
call has been recovered, it is evident that carrier traffic is at a low ebb." 
In the next day's summary appeared the last mention of carriers before 
December 7, and it was rather negative : "No information on submarines 
or carriers." 
Other messages, however, clearly indicated the drive to the south, 
which Japan made no attempt to conceal. Twice before, Rochefort, 
Fabian, Layton, and O.N.I, had seen exactly the same conditions , and 
twice before their reasoning that the carriers were being held in empire 
waters had been proved right. Now, they thought, they were seeing it 
happen again. Temporarily oblivious to the possibility of a surprise 
attack on Pearl Harbor, they watched the forces moving against Malaya 
as hypnotically as a conjuror's audience stares at the empty right hand 
while the left is pulling the ace out of a sleeve. 
 
American preconceptions were reinforced by two PURPLE messages of 
December 1, which the Navy read that same day. In the first, Tokyo 
directed Washington: "When you are faced with the necessity of 
destroying codes, get in touch with the naval attache's office there and 
make use of chemicals they have on hand for this purpose . The ATTACHÉ 
should have been advised by the Navy Ministry regarding this." Five days 
earlier, the cryptanalysts had read Tokyo's detailed instructions on how 
to destroy the PURPLE machine in an emergency. These two code-
destruction messages appeared to be just precautionary measures in a 
tense situation, and this impression was strengthened by the second 
message of December 1. It seemed to virtually announce a Japanese 
invasion of British and Dutch possessions and to relegate conflict with 
the United States to a subsequent date: "The four offices in London, 
Hong - kong , Singapore and Manila have been instructed to abandon the 
use of the code machines and to dispose of them. The machine in 
Batavia has been returned to Japan. Regardless of the contents of my 
circular message #2447 [which MAGIC did not have], the U.S. (office) 
retains the machines and the machine codes." American officials 
breathed easier. The messages appeared to give the United States a bit 
more of what it needed most—time, time to build up its pitifully weak  
Army and Navy. 
 
While the world gazed with tunnel vision toward Southeast Asia, and 
American radio intelligence envisioned the Japanese carriers in home 
waters, six of them—Akagi, Kaga, Hiryu, Soryu, Shokaku, and Zuikaku—
were in fact butting eastward through the high winds and waves of the 
vacant sea. Late in the afternoon of December 2, Tokyo time, the force 
picked up, apparently on a blanket broadcast, an electrifying open-code 
message intended for it: NIITAKA-YAMA NOBORE (" Climb Mount Niitaka"). It 
informed the strike force that the decision for war had been made and 
directed it to proceed with attack. Niitaka-yama, also known as Mount 
Morrison, is a peak on Formosa whose 12,956- foot elevation made it the 
highest point of what was then the Japanese empire. The symbolism 
could not have been lost on the officers. The force refueled from its 
tankers. 
 
Earlier that day, the Japanese consulate in Honolulu had received 
Circular #2445 in J19, relayed by Washington from Tokyo: 
 
Take great pains that this does not leak out. You are to take the 
following measures immediately: 
1.  With the exception of one copy each of the O [PA-K2] and the 
L [LA] codes, you are to burn all telegraph codes (this includes the 
codebooks for communication between the three departments 
[HATO] and those for use by the Navy). 
2.  As soon as you have completed this operation, wire the one 
word  HARUNA . 
3.  Burn all secret records of incoming and outgoing telegrams. 
4. Taking care not to arouse outside suspicion, dispose of all 
secret documents in the same way. 
Since these measures are in preparation for an emergency, keep 
this within your consulate and carry out your duties with calmness 
and care. 
 
The codes were duly burned, including the TSU, or J19, in which the 
circular was transmitted. That evening Kita sent HARUNA.  Henceforth  the  
consulate  code  secretary, Samon Tsukikawa, would have to transmit 
the spy messages of Yoshikawa, alias Morimura, in the simpler PA-K2. 
The first such message arranged four signaling systems by which a spy 
might report on the condition of the ships in Pearl Harbor. The 
arrangement had been submitted to Yoshikawa by an Axis spy in Hawaii, 
Bernhard Julius Otto Kühn.  Nazi Propaganda Minister Josef Goebbels 
had transferred him to the islands in 1935 after a contretemps with 
Kühn's daughter Ruth, who had become Goebbels' mistress when she 
was 16. In his signaling system, Kühn stipulated that numbers from 1  
to 8 would mean such things as A number of carriers preparing to sortie 
(which was 2) and Several carriers departed between 4th and 6th (which 
was 7). Then he arranged that bonfires, house lights shown at certain 
times and places, or want ads broadcast over radio station KGMG would 
mean certain numbers. For example, 7 would be represented by two 
lights shown in the window of a house on Lanikai Beach between 2 and 3 
a.m., or by two sheets between 10 and 11 a.m., by lights in the attic  
window of a house in Kalama between 11 and 12 p.m., or by a want ad 
offering a complete chicken farm for sale and listing P.O. Box 1476. If all 
these failed, a bonfire on a certain peak of Maui Island between 8 and 9 
p.m. would indicate 7. The purpose of the system was to eliminate 
dangerous personal contacts between Kühn and the Japanese. Kühn 
tested it on December 2, found that it worked, and passed it to 
Yoshikawa. He had it encoded  (in PA-K2)   and sent to Tokyo in two long 
parts on December 3. 
It was now the third day of the month in which the Japanese 
consulate gave its cable business to R.C.A. Following Sarnoff's 
instructions, George Street , district manager of the firm , had had the 
Japanese consulate messages copied on a blank sheet of paper with no 
identification of the sender or addressee. About 10 or 11 a.m., December 
3, Mayfield called at the branch office and Street slipped him a blank 
envelope containing the messages. As soon as Mayfield returned to the 
District Intelligence Office, he had a messenger bring them down to 
Rochefort. 
In Washington that Wednesday, the Signal Intelligence Service solved 
a PURPLE message from Tokyo—and the readers of MAGIC, who only two 
days earlier had been lulled by the supposition that Japan might 
temporarily spare the United States, were stunned by the realization that 
the arrow of war might be loosed momentarily. For the message ordered 
the Washington embassy to "burn all [codes] but those now used with 
the machine and one copy each of o code [PA-K2] and abbreviating code 
[LA]. . . . Stop at once using one code machine unit and destroy it 
completely . . . wire . . . HARUNA." Under Secretary of State Welles saw it 
and felt that "the chances had diminished from one in a thousand to one 
in a million that war could then be avoided." When the President's naval 
aide, Beardall, brought the message to Roosevelt, he said in substance, 
"Mr. President, this is a very significant dispatch." After the Chief 
Executive had read it carefully, he asked Beardall, "When do you think it 
will happen?"—referring to the outbreak of war. "Most any time," replied 
the naval aide, who thought that the moment was getting very close . 
At the Japanese embassy at 2514 Massachusetts Avenue, the code 
clerks were executing these destruction orders. The code room stood at 
the southeast corner of the embassy, with windows overlooking the 
embassy parking lot and another legation next door. Half a dozen desks 
clustered in the middle of the room. Two cipher machines waited on 
desks against the west wall and a third, broken, rested in the walk -in 
safe . In utter disregard of the regulations promulgated for the security of 
communications, the embassy had hired an elderly Negro janitor named 
Robert to dust and clean the code room and its supersecret furnishings 
each day. The code clerks did make some obeisance to the security 
regulations by not allowing him in the room unless some Japanese were 
in it. But the situation was, to say the least, ironical. While the Japanese 
Foreign Office was exercising almost superhuman security precautions 
and American cryptanalysts were suffering nervous breakdowns to solve 
the PURPLE machine, an American citizen was running his duster over 
tables on which stood the intricate machines that were the vortex of this 
silent struggle . 
But just as the Japanese seemed not to have given serious thought to 
the possibility of Robert's being a spy, so the Americans seemed to have 
given no serious thought to the possibility that a spy might have been 
insinuated into the Japanese embassy to ease their cryptanalytic burden. 
Of course, even if they had thought about it, they might have rejected the 
idea , for discovery of the spy would have meant an automatic change of 
codes. The danger of this was much less if the systems were read 
through cryptanalysis. 
The paper codes of the Japanese consisted of folders whose four or six 
pages could be opened into a single long sheet. Embassy Counselor 
Sadao Iguchi , who was in charge of the code room, directed telegraph 
officer Masana Horiuchi and code clerks Takeshi Kajiwara, Hiroshi Hori, 
Juichi Yoshida, Tsukao Kawabata and Kenichiro Kondo in the burning of 
the paper codes. Demolition of the code machine was more complicated, 
and followed the guidelines transmitted recently by the Foreign Office. 
The machines were dismantled with a screwdriver, hammered into 
unrecognizability, and then dissolved in acid from the naval attache's 
office to destroy them thoroughly. Some of these operations were carried 
out in the gardens of the embassy; so when Bratton, who had read the 
code-destruction intelligence, sent an officer to the embassy to check, he 
obtained immediate confirmation. 
Now the American officials realized the ominous meaning of the 
HARUNA messages that had been intercepted as they were sent from New 
York, New Orleans , and Havana and that had been received just that day 
in S.I.S. The Army and Navy high command universally regarded the 
destruction of codes as virtual certainty that war would break out within 
the next few days. As Stark's deputy put it: "If you rupture diplomatic 
negotiations you do not necessarily have to burn your codes. The 
diplomats go home, and they can pack up their codes with their dolls 
and take them home. Also, when you rupture diplomatic negotiations 
you do not rupture consular relations. The consuls stay on. Now, in this 
particular set of dispatches they not only told their diplomats in 
Washington and London to burn their codes, but they told their consuls 
in Manila, in Hong Kong, Singapore, and Batavia to burn their codes and 
that did not mean a rupture of diplomatic relations; it meant war." 
A few hours after the code-destruction MAGIC reached Stark, he 
dispatched the electrifying news to Kimmel and Hart : 
 
Highly reliable information has been received that categoric and 
urgent instructions were sent yesterday to Japanese diplomatic 
and consular posts at Hongkong X Singapore X Batavia X Manila X 
Washington and London to destroy most of their codes and ciphers 
at once and to burn all other important confidential and secret 
documents X 
 
He followed this five minutes later with another message: 
 
Circular twenty four forty four from Tokyo one December 
ordered London X Hongkong X Singapore and Manila to destroy 
PURPLE machine XX Batavia machine already sent to Tokyo XX 
December second Washington also directed destroy PURPLE X all but 
one copy of other systems X and all secret documents XX British 
Admiralty London today reports embassy London has complied 
 
In Washington urgency drove out all thoughts of security. The strict 
injunction against ever mentioning MAGIC was completely overlooked. 
When Kimmel got the message, he asked Layton what "PURPLE" was. So 
tight had security been that neither of them knew. They checked with 
Lieutenant Herbert M. Coleman, the fleet security officer, who told them 
that it was a cipher machine similar to the Navy's. 
 
At 8:45 p.m. that night, Thursday, December 4, the watch officer of 
the F.C.C.'s Radio Intelligence Division telephoned the Office of Naval 
Intelligence to ask if it could accept a certain message. The O.N.I. officer 
was not sure and said he would call back. At 9:05 GY watch officer 
Brotherhood called the F.C.C. and was given a Japanese weather report 
that sounded like something the F.C.C. man had been told to listen for. 
He read it to Brotherhood: "Tokyo: today—wind slightly stronger, may 
become cloudy tonight ; tomorrow-—slightly cloudy and fine weather. 
Kanagawa prefecture: today—north wind cloudy; from afternoon—more 
clouds. Chiba prefecture: today—north wind clear, may become slightly 
cloudy. Ocean surface: calm." 
Brotherhood was relieved that it included nothing about EAST WIND 
RAIN, which would have meant the United States, but in any case this 
message seemed to lack something that would have been required in a 
true execute. For one thing , the phrase NORTH WIND CLOUDY, which would 
have meant Russia, was not repeated twice. Nevertheless, Brotherhood 
telephoned Rear Admiral Leigh Noyes, director of naval communications, 
who remarked that he thought the wind was blowing from a funny 
direction. The concensus was that it was not a genuine execute, and the 
search continued. 
 
In Tokyo, where it was December 5, Foreign Minister Togo received 
representatives of the Army and Navy general staffs. A general and an 
admiral wanted to discuss the delicate matter of the precise timing of 
Japan's final note to the United States. Drafted in English by the director 
of the Foreign Office's American bureau, the note had been approved by 
the Liaison Conference, a six-man war cabinet, at its meeting the day 
before. It rejected Hull's offer of the 26th and concluded: "The Japanese 
Government regrets to have to notify hereby the American Government 
that in view of the attitude of the American Government it cannot but 
consider that it is impossible to reach an agreement through further 
negotiations." 
Article I of the 1907 Hague Convention governing the laws of war 
provides that ". . . hostilities . . . must not commence without previous 
and explicit warning, in the form either of a reasoned declaration of war 
or of an ultimatum with conditional declaration of war." Togo had 
suggested to the Liaison Conference that the note was far stronger than 
an ultimatum and that to include a specific declaration of war would be 
"merely to reiterate the obvious." The conferees had gratefully acceded to 
this casuistry, since it enabled them to comply with the prior -notification 
requirement without endangering the surprise of the attack. Since the 
Hague Convention does not specify how long in advance such notification 
must be given, Premier Tojo and the other conferees thought to shave the 
time as much as possible. Dawn in Hawaii was about noon m 
Washington. The Liaison Conference had tentatively set 12:30 p.m., 
Sunday, December 7 (Washington time), as the time of delivery of the 
note. 
But when the two military men called upon Togo the next day to fix 
the exact time, Vice Admiral Seiichi Ito, vice chief of the naval general 
staff, told the foreign minister [Togo later wrote] "that the high command 
had found it necessary to postpone presentation of the document thirty 
minutes beyond the time previously agreed upon, and that they wanted 
my consent thereto. I asked the reason for the delay , and Ito said that it 
was because he had miscalculated. ... I inquired further what period of 
time would be allowed between notification and attack; but Ito declined 
to answer this, on the plea of operational secrecy. I persisted, demanding  
assurance that even with the hour of delivery changed from twelve-thirty 
to one there would remain a sufficient time thereafter before the attack 
occurred; this assurance Ito gave. With this—being able to learn no 
more—I assented to his request. In leaving, Ito said: 'We want you not to 
cable the notification to the Embassy in Washington too early.'" In this 
demand lay the seeds of Japan's juridical culpability. 
 
Yoshikawa, in Honolulu, had continued sending his ship-disposition 
reports after the switch to PA-K2. They were an odd melange of accuracy , 
error , and outright falsehoods. On December 3, for example, he correctly 
reported that the liner  Lurline had arrived from San Francisco but stated 
that a military transport had departed when no such thing had occurred. 
The next day he informed Tokyo about the hasty departure of a cruiser of 
the Honolulu class; no such ship either entered or cleared the harbor on 
the 4th. Then, on the 5th, he cabled that three battleships had arrived in 
Pearl Harbor, making a total—which he reported with deadly accuracy—
of eight anchored in the harbor. His messages, sent over Kita's signature, 
were decoded in the Foreign Office and routed to the North American 
section, where Toshikazu Kase passed them immediately to the Navy 
Ministry. Here they were redrafted, encoded in a naval code, and 
transmitted on a special frequency not normally used by the Navy and 
without any direct address to the Pearl Harbor strike force. Commander 
Koshi decoded it and brought to his chief this latest information. 
The communication-security precautions paid off. Whether or not the 
messages slipped by the American radio monitors in Hawaii mattered 
little. Mere interception would not have helped much. The messages bore 
no external indication of their intended recipient, and they could not 
have been read. Rochefort's attack on Japanese naval codes had 
achieved some minor successes in late October and November, but he 
could read only about 10 per cent of the naval traffic, and much of this 
consisted of weather and other minor systems. The information obtained, 
Rochefort said, "was not in any sense vital." Cavite was spottily reading 
JN25 messages—which revealed nothing about Pearl Harbor—until 
December 4, when the superencipherment was suddenly changed. As a 
message that moved on the monitor channel put it: "Five numeral 
intercepts subsequent to zero six hundred today indicate change of 
cipher system including complete change differentials and indicator 
subtracters X All intercepts received since time indicated checked against 
all differentials three previous systems X No dupes." Corregidor was not 
to get the initial break into the new superencipherment until December 
8. And the only other system in which the Yoshikawa messages might 
have been forwarded—the flag officers' system—remained unsolved. 
A possibility of warning was opened at the source, however, when 
Yoshikawa's original messages became available to Rochefort's unit. 
Mayfield had picked up another batch of cables in the surreptitious 
fashion from Street on Friday morning and immediately sent them down 
to Rochefort's unit by messenger. Solving them was not part of its duty,4 
but when a superior officer and colleague asks one to do a favor , it is 
hard to say no. Rochefort assigned the messages to Chief Radioman 
Farnsley C. Woodward, 39, who had had some experience with Japanese 
diplomatic codes at the Shanghai station from 1938 to 1940. He had 
some help from Lieutenant Commanders Thomas H. Dyer, Rochefort's 
senior cryptanalyst, and Wesley A. Wright, Dyer's assistant. Although the 
unit was not working on the diplomatic systems, it had information on 
them in the Navy's R.I.P.s, or Radio Intelligence Publications, with which 
all radio intelligence units were supplied. The R.I.P. gave, however, only 
the PA code list, leaving the onerous reconstruction of the current K2 
transposition to the cryptanalyst. The half-dozen or so dispatches, plus  
some in LA, reached Woodward about 1:30 or 2 p.m. Friday, and he 
immediately began the first of a series of 12- and 14-hour days to read 
them. He had no difficulty with the LA messages, which were translated 
into English by Marine Corps Captain Alva Lasswell, but these yielded 
"nothing but junk ." The K2, however, eluded him, and he worked on it far 
into the night. 
 
In Tokyo it was a little after 1 p.m. on Saturday, December 6. The 
Japanese reply to Hull's note of the 26th had recently been sent to the 
cable room of the Foreign Ministry for transmission to the embassy in 
Washington. Kazuji Kameyama, the cable chief, broke it into fourteen  
approximately equal parts to facilitate handling and ordered these 
enciphered on the 97-shiki O-bun In-ji-ki. He also enciphered a shorter 
" pilot " message from Togo alerting the embassy that the reply was on the 
way and instructing it "to put it in nicely drafted form and make every 
preparation to present it to the Americans just as soon as you receive 
instructions." At 8:30 p.m., the pilot message was telegraphed from the 
cable room to Tokyo's Central Telegraph Office, from where, 45 minutes 
later, it was radioed to the United States. Bainbridge Island intercepted it 
and relayed it to OP-20-G. By five minutes past noon on Saturday, 
December 6 (Washington time), OP-20-o had delivered the teletype copy 
to S.I.S., which promptly ran it through the PURPLE machine. By 2 p.m. 
Bratton had it, translated and typed. An hour later it was in the hands of 
the Army distributees. S.I.S. had officially closed at 1 p.m. and was not 
due to reopen until 6, when it was to go on 24-hour status. But this 
notification of the imminent receipt of the long-awaited reply to Hull's 
note of the 26th led to telephoning employees Mary J. Dunning and Ray 
Cave about 2:30 and asking them to report to work. By 4 both were 
there. 
In Tokyo, Kameyama had released the first 13 parts of the Japanese 
note to the Central Telegraph Office. Following the instructions of the 
American bureau, he retained the crucial 14th part, which broke off 
negotiations. Shortly after 10 p.m., commercial radio began sending the 
13 parts to Washington. Most of them took less than ten minutes to 
transmit, but even though two transmitters were used, it was not until 
two minutes before 2 a.m. that the tail of the last part had gone . 
Bainbridge, of course, was listening, and it picked the parts up in this 
order: 1, 2, 3, 4, 10, 9, 5, 12, 7, 11, 6, 13, 8. One batch arrived by 
teletype at OP-20-G at eleven minutes before noon, Saturday, December 
6, Washington time, and the other at nine minutes of 3 that afternoon. 
Though it was Saturday, December 6, an even date and hence an Army 
date of responsibility, the Navy handled the dispatches because it knew 
that S.I.S. was not expected to work that afternoon, and it considered the 
intercepts of great importance. Decryptment did not go very smoothly, 
however. Something seemed to be in error. GY knew the key, but it was 
producing garbles every few letters. The cryptanalysts tried to correct 
them. 
Meanwhile, a decode into Japanese of the long PA-K2 message that 
Yoshikawa had sent concerning Kuhn's visual-signal system for Hawaii 
was placed on the desk of Mrs. Edgers in GZ. "At first glance," she said, 
"this seemed to be more interesting than some of the other messages I 
had in my basket , and so I selected it and asked one of the other men, 
who were also translators working on other messages, whether or not 
this shouldn't be done immediately and was told that I should and then I 
started to translate it. Well, it so happened that there was some mistake 
in the message that had to be corrected and so that took some time. That 
was at 12:30 or perhaps it was a little before or after 12:30; whatever 
time it was, we were to go home. It being Saturday, we worked until 
noon. I hadn't completed it, so I worked overtime and finished it, and I 
would say that between 1:30 and 2 was when I finished my rough draft  
translation." Mrs. Edgers left it in the hands of Chief Yeoman Bryant. But 
the message was still not entirely clear, and she had not yet had enough 
experience for her translations to be sent out without further checking. 
Kramer, busy with the 13 parts, did not examine it in detail. 
To speed processing of the 13 parts, GY, learning that some people 
were in S.I.S., sent over parts 1 and 2. But when Major Doud of S.I.S. 
ordered Miss Cave to OP-20-G to help in the smooth typeups, the two 
parts were returned to GY for solution there, probably because of the 
garbles. But other messages also coming in were retained by S.I.S. 
At 3 o'clock, Kramer, in GZ, had checked with GY to find out whether 
any more Tokyo traffic had come in before releasing his translators for 
the day. Since the critical matter of a diplomatic note is often found in 
the last sentences , GY broke down the last part intercepted for him. The 
first part of the first line indicated in Japanese that this was part 8 of a 
14-part message. After about three lines of Japanese text in the 
preamble, the message came out in English, just as the Foreign Office 
had sent it. Kramer could let his translators go home. Interspersed 
throughout the English text were many of the three-letter codewords 
indicating punctuation, paragraphing, and numbering, but these posed 
no problem since they had been recovered long ago. 
At 4 o'clock, when Linn took over the GY watch, the garbles still had 
not been cleared. He decided to start from the very beginning, to check 
the key, find what was wrong , and redecrypt the messages rather than to 
try to guess at the garbled letters and possibly make serious errors that 
would distort the sense. Discarding all the previous work caused a 
serious jam on the Navy's one PURPLE machine, and about 6 p.m. GY 
again called on S.I.S. for help. Parts 9 and 10 were sent over; an hour 
later, the decrypts came back in longhand. By 7:30, the last of the 13 
parts was being decrypted. 
Not all the garbles had been scrubbed out. Part 3 had a 75-letter 
smudge that could not be read at all, Part 10 a 45-letter blur , and Part 
11 one of 50 letters. Part 13 went awry in two patches. One deciphered 
as andnd and the other as chtualylokmmtt; GY thought the first should be 
and as and the second China, can but.5
In the Japanese embassy, about a mile away, the code clerks had 
completed deciphering the first seven or eight parts of the message by 
dinnertime. Then they all repaired to the Mayflower Hotel for a farewell  
dinner for Hidenari Terasaki, head of Japanese espionage for the western 
hemisphere, who had been ordered to another post. 
 
While they were enjoying themselves, American code clerks at the 
Department of State were at work encoding a personal appeal for peace 
from the President of the United States to the Emperor of Japan. This 
had been off again, on again since October, Roosevelt apparently wishing 
to save it for a last resort . Now he decided that the time had come. The 
message was on its way by 9 o'clock. It traversed the 7,000 miles to 
Tokyo in an hour. 
But it took ten hours to get from the Central Telegraph Office to the 
American embassy. 
 
As the President was addressing a message of peace to the Emperor, 
the men of the Japanese strike force were listening to a message of war. 
Shortly before, Admiral Nagumo had topped off the fuel tanks of his 
combat ships for the final dash. His crews waved farewell to the slow-
moving tankers. Now the officers read a stirring message from Yamamoto 
to all hands: "The moment has arrived. The fate of the empire is at stake . 
Let every man do his best." Banzais rent the air. Up the mast of Akagi 
fluttered the very flag that had flown at Japan's great naval victory over 
Russia in 1905. It was a moment of great emotion. Nagumo altered 
course to due south and bent on 26 knots. Through a mounting sea, the 
battle force plunged toward its target. 
 
Lovely, peaceful, that target lay "open unto the fields, and to the sky," 
oblivious to the onrushing armada of destruction. And as it increased its 
speed, more information for its mission was starting on its way. The 
R.C.A. office was time-stamping "1941 Dec 6 PM 6 01" on a message from 
the consulate. It was signed "Kita" but it came from Yoshikawa. It was 
brief (only 44 groups) and cheap ($6.82), but it reported that "(1) On the 
evening of the 5th, the battleship Wyoming and one sweeper entered 
port. Ships at anchor on the 6th were: 9 battleships, 3 minesweepers, 3 
light cruisers, 17 destroyers. Ships in dock were: 4 light cruisers, 2 
destroyers. Heavy cruisers and carriers have all left. (2) It appears that 
no air reconnaissance is being conducted by the fleet air arm." 
Yoshikawa was, as usual, partly right and partly wrong. He mistook Utah 
for Wyoming. His figure on the battleships was correct, but in harbor that 
afternoon were 6 light and 2 heavy cruisers, 29 destroyers, 4 
minesweepers, 8 minelayers, and 3 seaplane tenders. With this message 
Yoshikawa completed his assignment. It was the last cable sent by the 
Japanese consulate in Hawaii for many years. 
 
[Codebreakers 050.jpg]
On the eve of Pearl Harbor, Takeo Yoshikawa  sends  his final message over Consul 
Kita's signature, using the PA-K2 code, to report that the U.S. fleet is still in port 
 
By 8:45 p.m. in Washington, the 13 parts had been typed in smooth 
copies and put up in folders. Kramer began telephoning the recipients to 
find out where they were so he could bring the MAGIC to them. He also 
called his wife , Mary, who agreed to chauffeur him during his deliveries. 
They reached the White House first, at about 9:15. The naval aide, 
Beardall, had told the President that some MAGIC would be delivered that 
evening, and at about 4 p.m. he had ordered his communications 
assistant, Lieutenant Lester R. Schulz , to stand by and bring it to the 
President. Schulz was waiting in Beardall's small office in the corner of 
the basement mail room in the White House when Kramer arrived. The 
Roosevelts had been entertaining at a large dinner party, but the 
President had excused himself. Schulz obtained permission to bring the 
MAGIC to the President, and an usher accompanied him to the oval study 
on the second floor and announced him. Roosevelt was seated at his 
desk. Only Harry Hopkins was with. him. Schulz unlocked the briefcase 
with the key that Beardall had given him, removed the sheaf of MAGIC, 
and handed it to the President. He read the 13 parts in about ten 
minutes while Hopkins paced slowly up and down. Then Hopkins read 
them. The  13th part rejected Hull's offer, and when Hopkins had passed 
the papers back to the President, Roosevelt turned to him and said, in 
effect, "This means war." Hopkins agreed, and for about five minutes they 
discussed the situation, the deployment of Japanese forces, the 
movement towards Indochina, and similar matters. The President 
mentioned his message to Hirohito . Hopkins remarked that it was too 
bad that the United States could not strike the first blow and prevent any 
kind of surprise in the inevitable war. 
"No," the President said in effect, "we can't do that. We are a 
democracy and a peaceful people." He raised his voice: "But we have a 
good record." He tried unsuccessfully to get Admiral Stark on the 
telephone, deciding against having him paged at the National Theater for 
fear of causing undue alarm . 
The President then returned the papers to Schulz and, about half an 
hour after he had entered the study, Schulz left. He found Kramer seated 
at one of the long tables in the mail room. Schulz gave him the pouch  
and soon thereafter went home. Kramer, however, continued to the 
Wardman Park Hotel, where Secretary Knox had a suite . For about 
twenty minutes, while Kramer chatted with Mrs. Knox and the acting  
manager of Knox's Chicago Daily News, the Secretary read the 13 parts. 
He agreed with Kramer, that, even incomplete, it pointed to a termination 
of negotiations. He went into another room to make some telephone calls, 
and when he came out he told Kramer to bring the latest MAGIC to a 
meeting that had been arranged for 10 a.m. the next morning with 
Stimson and Hull in the State Department. (Bratton had delivered the 13 
parts to the night duty officer at State at 10 p.m., admonishing him to 
get them to Hull at once.) Knox returned the intercepts to Kramer, who 
then went to the home of Rear Admiral Theodore S. Wilkinson , director of 
naval intelligence, where Beardall and Army intelligence chief Brigadier  
General Sherman Miles happened to be dinner guests. All three studied 
the intercept in a room away from the other guests, Beardall reading 
from an extra copy that Kramer had. They too seemed to feel that 
negotiations were coming to an end. 
It was after midnight when Kramer left the Wilkinson house. His wife 
drove him back to the Navy Department, where he put the MAGIC back in 
his safe in GZ and checked to see if the 14th part had yet come in. It had 
not. Finally he went home himself. 
In S.I.S., meanwhile, the new teletype that would expedite the 
forwarding of intercepts was being set up in the " cage ," the barred room 
where PURPLE traffic was processed. Monitor Post 2 was requested to send 
in some intercepts as a test. In San Francisco, Harold W. Martin, the 
noncom in charge, punched onto the teletype tape the intercepts that the 
post had picked up since airmailing in the bulk of the day's material, as 
well as the earlier ones. Among the later ones was Yoshikawa's final 
message, which thus became one of the first to move on the direct wire 
as a real, nontest item. S.I.S. received it a little after midnight. But PA-K2 
was a low-priority system, and the message had originated in a consular 
office. It was set aside to be worked on later. 
Besides , S.I.S. had more important things to worry about. Like OP-20-
O, it was going frantic in a search for the 14th part. Captain Robert E. 
Schukraft, head of the intercept section, and Frank B. Rowlett, the 
civilian cryptanalyst in charge of the Japanese diplomatic solutions, 
checked and rechecked to see whether one of the stations had picked it 
up and had somehow neglected to forward it. The message preambles 
had said that it existed, but they could find no trace of it. Neither 
suspected that the Japanese Foreign Office had deliberately held up 
transmission of this final conclusive part for security's sake. 
Neither did the code clerks at the Japanese embassy. They had 
returned from Terasaki's party about 9:30, and by midnight had 
completed deciphering of the 13 parts. While they waited for the final 
section, they busied themselves by disposing of the remnants of the 
cipher machine they had destroyed the night before. But they did 
nothing to fulfill the orders of the pilot message to prepare the dispatch 
for immediate presentation. 
Finally, fourteen hours after the last part of the previous 13 parts had 
been transmitted, the Foreign Office released the crucial 14th part that 
broke off negotiations. At 4 p.m., Tokyo time, it ordered it transmitted via 
both R.C.A. and Mackay Radio & Telegraph Company to ensure its 
correct reception . An hour and a half later, it wired to the Central 
Telegraph Office the coded message ordering the 1 p.m. delivery of the 
14-part note. This too was sent via the two companies. 
As usual, the indefatigable ear of Bainbridge Island detected the 
ethereal pulses of both messages. It picked up the Mackay transmission 
of the 14th part between 12:05 and 12:10 a.m., December 7, local time, 
and the even briefer one o'clock message between 1:28 and 1:37 a.m. It 
teletyped them to GY in a single transmission, the 14th part as serial No. 
380 of Station S, the one o'clock as No. 381. Brotherhood, who was GY 
watch officer, ran them through the PURPLE machine. He evidently had 
some trouble with the 14th part, for it took an hour to break. But by 4 
a.m. he had it in English. The three-letter codegroups were quickly 
translated into punctuation; the message would need little more than 
typing. The one o'clock message, however, turned out to be in Japanese. 
He sent it to S.I.S. for translation, knowing that translators were on duty 
because S.I.S. was beginning its round-the-clock tours. It was a little 
past 5 a.m., Washington time. 
In the embassy of Nippon, the code clerks who had waited all through 
the night for the 14th part were, on Counselor Iguchi's advice, being sent 
home. Just as they were climbing wearily into their beds, the naval 
attaché arrived and found the mailbox stuffed with cablegrams. The duty 
officer telephoned the clerks at their homes about 8 a.m. and ordered 
them back to work. 
 
A few hundred miles north of Oahu, the Japanese task force, bristling 
with guns , planes, and hate for Americans, bore down on the Pacific 
Fleet. A few hours earlier, a message had arrived from Tokyo that caused 
Commander Mitsuo Fuchida, the pilot who was to head the first wave of 
the air attack, to breathe a sigh of relief. It had been relayed from 
Yoshikawa, and it reported that no barrage balloons had yet been 
emplaced to protect the fleet from air attack. The same message also 
caused Commander Minoru Genda to sigh with relief. It stated that the 
battleships appeared not to be protected by torpedo nets . Genda had 
conceived the plan of shallow-water torpedo attack on the anchored 
American ships. 
A little more than an hour after the hands of Honolulu clocks had 
snipped off December 6 and opened out into the first hours of December 
7, the Pearl Harbor strike force received Tokyo's relay of Yoshikawa's 
final message. The American ships were still in harbor, awaiting the ax 
stroke with fat complacency. They were apparently not even protected by 
air search. Was it all a decoy? The strike force's radio officer, Commander 
Kanjiro Ono, listened intently to Honolulu's radio station KGMB for any 
inkling that the Americans knew of them. He heard only the soft  
melodies of the islands. On Hiryu, the flight deck officer slipped bits of 
paper between each plane 's radio transmitter key and its contact point to 
make sure that radio silence, so carefully preserved for almost two 
weeks, would not be accidentally broken in the last few hours to destroy 
the element of surprise. 
 
As Yoshikawa's final report was being decoded aboard Akagi, Kramer 
returned to the Navy Department he had left only seven hours before, 
and began working again. It was 7:30 on the morning of Sunday, 
December 7. 
Brotherhood's decryptment of the 14th part was on his desk when he 
arrived. It took him about half an hour to ready a smooth version, and at 
8 o'clock he delivered the neatly typed copy to McCollum. Other copies 
went to S.I.S. for its distribution. Kramer then worked on other traffic in 
his office, interrupting himself only once, at 8:45, to bring a copy of the 
14th part to naval intelligence chief Wilkinson on his arrival at the Navy 
Department. At 9:30 he set out to deliver the full 14 parts to the meeting 
of the three secretaries. He stopped at the office of the Chief of Naval 
Operations to make sure that Stark had been given the message, which 
he had, and then walked and trotted to the White House. He got there at 
about 9:45 and gave the MAGIC pouch to Beardall, who had assigned 
himself to duty that morning because he thought the 14th part of the 
message that be had seen at Wilkinson's house the night before might be 
coming in. 
Beardall brought the folder to the President, who was 
in his bedroom. Roosevelt said good morning to him, read the 
intercept, and commented that it looked like the Japanese were going to 
break off negotiations. Then he returned the MAGIC, and Beardall took it 
back to the Navy Department. 
Kramer, meanwhile, had hurried across the west lawn of the White 
House to the ugly , ornate State Department building, arriving at about 
ten minutes of 10. The Army courier appeared at almost the same 
moment with the MAGIC for Hull and Stimson. Three State Department 
officials who saw MAGIC—Hornbeck, Ballantine, and Hamilton —were 
shown the 14th part by Hull's aide, John Stone , and the group discussed 
the situation in general terms until the secretaries arrived a few minutes 
later. Kramer gave his pouch to Knox and headed back to the Navy 
Department. 
Meanwhile, the translation of the one o'clock message had come up 
from S.I.S. It was placed in Bratton's hands about 9 a.m. while he was 
reading the 14th part. It "immediately stunned me into frenzied activity 
because of its implications, and from that time on I was busily engaged 
trying to locate various officers of the general staff and conferring with 
them on the exclusive subject of this message and its meaning," he said 
later. He tried first to get in touch with Marshall, calling him at his 
quarters at Fort Myer, and was told by an orderly that the chief of staff 
had gone on his customary Sunday morning horseback ride . Bratton 
directed the orderly: 
"Please go out at once, get assistance if necessary, and find General 
Marshall, ask him to—tell him who I am and tell him to go to the nearest 
telephone, that it is vitally important that I communicate with him at the 
earliest practicable moment." The orderly said he would. Bratton called 
Miles, told him of the message, and urged him to come down to the office 
at once. Between 10 and 10:30, Marshall called Bratton back. The 
colonel offered to drive out at once with the one o'clock message, but 
Marshall told him not to bother, that he was coming down to his office at 
once. Bratton obeyed. 
Kramer arrived back in GZ at about 10:20, and found there the one 
o'clock message. It struck him as forcibly as it had Bratton. He at once 
had Yeoman Bryant prepare a new set of folders for immediate delivery of 
the intercept. Included in the new set were other messages which 
S.I.S. had decrypted, and on which Kramer had been working earlier 
in the morning: Tokyo serial No. 904, which directed the ambassadors 
not to use an ordinary clerk in preparing the 14-part ultimatum for 
presentation to the Secretary of State, so as to preserve maximum  
security; serial No. 909, thanking the two ambassadors for all their 
efforts; and serial No. 910, ordering destruction of the remaining cipher 
machine and all machine codes. 
Kramer was about to dart out again when Pering, the GY watch officer, 
brought in a message in plain-language Japanese, ending with the 
telltale STOP that indicated it was an INGO DENPO message: KOYANAGI 
RIJIYORI SEIRINOTUGOO 
ARUNITVKI HATTORI MINAMI KINEBUNKO SETURITU KIKINO 
KYOKAINGAKU SKYUU DENPOO ARITASI STOP TOGO. Kramer 
recognized KOYANAGI as the codeword for England, and HATTORI as a 
codeword whose meaning he did not recall. He consulted his code list 
and saw that it meant Relations between Japan and {name of country) are 
not in accordance with expectation. But in his haste he overlooked that 
the common Japanese word minami, which means "south," had an INGO 
DENPO meaning of U.S.A. He interpreted the message as "Please have 
director Koyagani send a wire stating the sum which has been decided to 
be spent on the South Hattori Memorial Library in order that this 
business may be wound up." Consequently, he dictated a decode that 
omitted United States: Relations between Japan and England are not in 
accordance with expectation. Yeoman Bryant inserted this and three 
other minor messages that had come over from the Army into the folders. 
Kramer meanwhile made a navigator's time circle that indicated that one 
o'clock in Washington was dawn in Hawaii and the very early hours of 
the morning in the Far East around Singapore and the Philippines, 
which everybody seemed to be watching. He shoved the folders into the 
briefcase and dashed out the door. 
He went first to Stark's office, where the officers were discussing the 
14th part, summoned McCollum, gave him the pouch that included the 
final code-destruction and one o'clock messages, and mentioned to him 
the significance of the latter's timing. McCollum grasped it at once and 
disappeared into Stark's office. Kramer wheeled and hurried down the 
passageway. He emerged from the Navy Department building and turned 
right on Constitution 
Avenue, heading for the meeting in the State Department three or four 
blocks away. The urgency of the situation washed over him again, and he 
began to move on the double. 
He half trotted, half walked to State, getting there at about 10:45. 
Hull, Knox, and Stimson were still meeting. Kramer saw them grouped 
around the conference table when the door to Hull's office was opened 
briefly . He gave the MAGIC messages to Stone, explaining to him how the 
one o'clock time of delivery of the ultimatum tied in with the movement of 
a big Japanese convoy down the coast of Indochina, and mentioning in 
passing that the time in Hawaii would be 7:30 a.m. The final code-
destruction message was self-explanatory. Kramer carried a MAGIC pouch 
to the White House, and then returned, perspiring, to the Navy 
Department, to busy himself with still more MAGIC. At about 12:30, he 
spotted the omission of United States from the INGO DENPO message. 
Because the one o'clock meeting was so close, he telephoned the 
recipients with the correction , a practice he had followed several times in 
the past, but reached only McCollum and Bratton. He told them that 
United States was to be inserted in file number 7148. The force of it had 
been considerably lessened by the one o'clock message, but Kramer, 
conscientious beyond the basic requirements of duty, nevertheless 
planned to send around a corrected version. 
Safford later estimated that OP-20-G handled three times as much 
material that weekend as on a normal one; the GY log shows at least 28 
messages in PURPLE alone handled that Sunday. And these messages 
were processed much more expeditiously than at any other time in the 
past, Kramer said. The cryptanalysts had done their duty, and had done 
it superbly. Events now passed out of their hands. 
 
In Tokyo, the President's message to the Emperor had finally been 
delivered to Grew after a delay of ten hours. The chief of the censorship 
office had ordered that all foreign cables be held up for five hours one 
day and ten hours the next. The order had been issued at the request of 
a lieutenant colonel on the general staff, who asked that this be done "as 
a precaution." The President's "triple priority" message arrived on one of 
the ten-hour days, was stalled for the required time, and was finally 
delivered at 10:30 p.m., Tokyo time. 
Grew immediately arranged for a meeting with Togo and, when the 
message had been decoded, drove to Togo's official residence at 12:15 
a.m. He requested—as is the right of all ambassadors—an audience with 
the head of state to present the message, then read it aloud to Togo and 
gave him a copy. Togo promised to present the matter to the Throne and, 
despite the lateness of the hour, telephoned the Lord Keeper of the Privy  
Seal for an audience. Ministers of state would be received at any hour, 
and the audience was arranged for 3 a.m. Togo began having the 
message translated. 
 
It was then about 5:30 a.m., December 7, in Hawaii. The Japanese 
task force was only 250 miles north of Pearl Harbor. More than 2,000 
Americans with less than three hours to live slept or played in blissful 
ignorance of that fact. The hands of clocks in the Foreign Office in Tokyo, 
in the code room at the Japanese embassy in Washington, in the War 
and Navy departments, in Pearl Harbor, circled around and around, but 
not so quickly as the spinning propellers of Nagumo's ships. At 5:30, two 
cruisers catapulted off a pair of scout planes to make sure the Americans 
were still there. 
 
The clerks at the embassy had straggled back to work between 9:30 
and 10. They began decoding the longer cables first, as experience had 
shown that these were usually the more important. At the same time, the 
embassy's first secretary, Katzuso Okumura, was typing up the first 13 
parts of the ultimatum. He had been chosen because the Foreign Office 
had forbidden the use of an ordinary typist in the interests of secrecy 
and he was the only senior official who could operate a typewriter at all 
decently. At about 11:30, code clerk Juichi Yoshida adjusted the 
Alphabetical Typewriter to the proper keys and typed out a short code 
message. To the consternation of the entire staff, it turned out to be an 
instruction to deliver the 14-part message to Secretary Hull at 1 p.m., 
Washington time. The 14th part had not even been decoded from the 
sheaf of incoming cables! And only one code machine was left to decipher 
all the messages! 
A few blocks away, General Marshall had just arrived at the War 
Department. On his desk was the MAGIC folder with the 14-part message 
on top and the one o'clock message under it. He began to read the 
ultimatum carefully, some parts several times. Bratton and Brigadier 
General Leonard T. Gerow, the war plans chief, tried to get him to look at 
the one o'clock message, but it is rather difficult for subordinates to 
interrupt a four- star general, and he finished the ultimatum before 
finding the time-of-delivery message. It struck him with the same sense 
of urgency that it had the others, and he picked up the telephone to call 
Stark to see if he wanted to join him in sending a warning message to 
American forces in the Pacific. 
At approximately the same time, Ambassador Nomura was calling 
Hull to request an appointment at 1 p.m. And 230 miles north of Hawaii, 
the first wave of Japanese planes was thundering off the flight decks of 
the carriers. 
Stark was at that moment discussing the significance of the one 
o'clock message with Captain R. E. Schuirman, Navy's liaison with State. 
He told Marshall that he felt that enough warnings had been sent and 
that more would just confuse the commanders. Marshall thereupon 
wrote out the dispatch he wanted sent: 
 
Japanese are presenting at one p.m. Eastern Standard Time 
today what amounts to an ultimatum also they are under orders to 
destroy their code machine immediately Stop Just what 
significance the hour set may have we do not know but be on alert 
accordingly Stop 
 
On his desk Marshall had a scrambler telephone with which he could 
have called Short in Hawaii. The scrambling apparatus stood in a room 
next to his office, thus obviating the possibility of tapping the 
conversation in unscrambled form, as was done in commercial cases. 
But Marshall knew that scramblers afforded protection merely against 
casual listeners; they could be penetrated by a determined eavesdropper 
with proper equipment. He had on several occasions warned the 
President about security on his transatlantic telephone conversations 
with Ambassador Bullitt in France and later with Churchill—a wise 
move, for, though he did not know it, the Nazis had already penetrated 
that scrambler. The Japanese had evidenced some interest in the San 
Francisco-Honolulu scrambler, and Marshall was acutely sensitive "that 
the Japanese would have grasped at most any straw" to suggest to the 
isola- 
tionists that the administration had committed an overt act that had 
forced the Japanese hand. Japanese interception of a scrambler warning 
might thus have sent the country to war divided . So Marshall shunned 
the scrambler telephone and relied on the slightly slower but much more 
secure method of enciphering a written message. 
As he was completing the message, Stark called him back. He had 
reconsidered and wanted Marshall to add the usual admonition to show 
the message to the naval opposites. Marshall added: "Inform naval 
authorities of this communication." Stark offered the Navy 
communication facilities, but Marshall said that the Army's could get the 
message out as quickly. 
Marshall gave the message to Bratton to take it to the War 
Department message center for transmission to the commanding 
generals in the Philippines, Hawaii, the Caribbean , and West Coast, after 
vetoing a suggestion that it be typed first. As Bratton was leaving, Gerow 
called out that if there was any question as to priority, to send it to the 
Philippines first. Bratton, greatly agitated, gave the message to Colonel 
Edward French in the message center and asked how long it would take 
to get it out. French told him that it would be encoded in three minutes, 
on the air in eight, and in the hands of the addressees in twenty. Bratton 
returned and reported to Marshall, who did not understand the 
explanation and sent him back for a clarification. He still was not sure 
and sent Bratton back a third time, after which he was finally satisfied 
with the answer. 
Meanwhile, French had had the message typed anyway and then 
ordered it encoded on a machine that was operated from a typewriter 
keyboard . During the few minutes that this took, he checked his 
Honolulu circuit, and found that since early morning interference had 
been so bad that the small 10- kilowatt War Department radio could not 
" bust " through it. He knew that R.C.A. in San Francisco had a 40-
kilowatt transmitter which would have no difficulty in getting through, 
and that Western Union in San Francisco had a tube running across the 
street from its office to this R.C.A. office. He had also learned on the 
previous day that R.C.A. was installing a teletype circuit from its office in 
Honolulu to Short's headquarters at Fort Shafter. French figured that 
this would therefore be his most expeditious route; after the message had 
been 
encoded, he personally carried it over to his bank of six Western 
Union teletypes and, at 12:01 p.m. December 7, sent it on its way. 
Western Union forwarded it at 12:17, and 46 minutes later it was 
received by R.C.A. in Honolulu. Local time was 7:33 a.m. The first wave 
of Japanese planes was then only 37 miles away—so close that the Army 
radar operators at Opana Point, who had tracked the flight for several 
hours and had been told to " Forget it" when they first reported it, were 
about to lose it in the dead zone of the nearby hills. But though the 
teletype connection for Fort Shatter had been completed the day before, 
it was not in operation pending tests on Monday. R.C.A. put Marshall's 
message in an envelope marked "Commanding General" for hand 
delivery. 
 
In Tokyo, Togo had been received by the Emperor. He read the text of 
Roosevelt's message, then a draft of the imperial reply that he and Tojo 
had prepared. It stated that the 14-part note was to be considered as 
Japan's response. Hirohito assented, and at 3:15 a.m. Togo withdrew 
from the Divine Presence. Deeply moved, he recalled, "I passed solemnly, 
guided by a Court official, down several hundred yards of corridors, 
stretching serene and tranquil. Emerging at the carriage entrance of the 
Sakashita Gate, I gazed up at the brightly shining stars , and felt bathed 
in a sacred spirit . Through the Palace plaza in utter silence, hearing no 
sound of the sleeping capital but only the crunching of the gravel 
beneath the wheels of my car, I pondered that in a few short hours would 
dawn one of the eventful days of the history of the world." Even as he 
pondered, Japanese planes were circling over Pearl Harbor. 
 
In stark contrast to the calm stillness of Tokyo was the hectic bustle 
of the Japanese embassy on Massachusetts Avenue. 
Soon after the one o'clock message had been decoded, Okumura 
finished typing the first 13 parts. But he decided that this rough draft 
did not suit the formality of a document to be delivered to the Secretary 
of State. He began retyping it from the very beginning, being assisted 
now by a junior interpreter, Enseki. His task was complicated by two 
messages sent up from the code room, one ordering the insertion of a 
sentence that had been accidentally 
dropped, one changing a word. This required the retyping of several 
pages, including one just completed with a great deal of trouble. At about 
12:30, the code room finally gave him the 14th part of the ultimatum, 
but Okumura was nowhere near finished with the first 13. Nomura kept 
poking his head in the door to hurry him on. A few minutes after one, 
when it was evident that the document would not be finished for some 
time, the Japanese called Hull to request a postponement to 1:45, saying 
that the document they wished to present was not yet ready. Hull 
acquiesced. 
At almost exactly the time that the call to Hull was being placed, 
Commander Fuchida and his flight of 51 dive bombers, 49 high-level 
bombers, 40 torpedo planes, and 43 fighters arrived over Pearl Harbor. 
He fired a "black dragon" from his signal pistol to indicate that the 
squadrons should deploy in the assault pattern for complete surprise. 
Nine minutes later, he wirelessed the message "To, to, to"—the first 
syllable of the Japanese word for "Charge!" and the signal to attack. As 
the planes moved into position for their runs, he felt so certain that he 
had achieved complete surprise that, at 7:53, two minutes before the 
first bomb even fell, he jubilantly radioed " TORA ! TORA! TORA!" (" Tiger ! Tiger! 
Tiger!")—the prearranged codeword that indicated surprise. On Akagi, 
Nagumo turned to a brother officer and grasped his hand in a long, silent 
handshake. At 7:55, the first bomb exploded at the foot of the seaplane 
ramp at the southern end of Ford Island in the middle of Pearl Harbor. 
Okumura was still typing. His fingers struggled with the keys as 
torpedoes capsized Oklahoma, as bombs sank West Virginia, as 1,000 
men died in the searing inferno of  Arizona . At 1:50 p.m. Washington 
time, 25 minutes after the attack had started, he reached the end of his 
typing marathon. The two ambassadors, who were waiting in the 
vestibule, started for the State Department as soon as it was handed to 
them. 
 
The Japanese envoys arrived at the Department at 2:05 and went to 
the diplomatic waiting room [Hull wrote]. At almost that moment the 
President telephoned me from the White House. His voice was steady but 
clipped. 
He said, "There's a report that the Japanese have attacked Pearl 
Harbor." 
"Has the report been confirmed?" I asked. 
He said, "No." 
While each of us indicated his belief that the report was probably 
true, I suggested that he have it confirmed, having in mind my 
appointment with the Japanese Ambassadors. . . . 
Nomura and Kurusu came into my office at 2:20. I received them 
coldly and did not ask them to sit down. 
Nomura diffidently said he had been instructed by his Government to 
deliver a document to me at one o'clock, but that difficulty in decoding 
the message had delayed him. He then handed me his Government's 
note. 
I asked him why he had specified one o'clock in his first request for an 
interview . 
He replied that he did not know, but that was his instruction. 
I made a pretense of glancing through the note. I knew its contents 
already but naturally could give no indication of this fact. 
After reading two or three pages, I asked Nomura whether he had 
presented the document under instructions from his Government. 
He replied that he had. 
When I finished skimming the pages, I turned to Nomura and put my 
eye on him. 
"I must say," I said, "that in all my conversations with you during the 
last nine months I have never uttered one worth of untruth. This is borne 
out absolutely by the record. In all my fifty years of public service I have 
never seen a document that was more crowded with infamous falsehoods 
and distortions— infamous falsehoods and distortions on a scale so huge 
that I never imagined until today that any Government on this planet  
was capable of uttering them." 
Nomura seemed about to say something. His face was impassive, but I 
felt he was under great emotional strain . I stopped him with a motion of 
my hand. I nodded toward the' door. The Ambassadors turned without a 
word and walked out, their heads down. 
The warlords' hopes of shaving the warning time to the closest 
possible margin had quite literally gone up in the smoke of attack, and 
Japan had 'started hostilities without giving prior notification. Later, this 
failure to declare war would be made part of the charges on which the 
Japanese war criminals were tried—and convicted, some of them paying 
with their lives . Togo would try to exonerate himself by throwing the 
blame on the embassy personnel for neglecting to decipher the cables 
promptly and to type the ultimatum at once. Perhaps some lawyer 's 
talking point might have been salvaged if the ambassadors had grabbed 
Okumura's original copy, no matter how messy, and taken it to Hull at 1 
p.m., or if they had taken the first few pages of the fair copy at 1 p.m. 
and directed the embassy staff to rush the other pages over as 
completed. But even if the entire document had been delivered on time, 
the 25 minutes that remained until the attack would not have been 
sufficient time for all the steps needed to prevent surprise: reading the 
document, guessing that a military attack was intended, notifying the 
War and Navy departments, composing, enciphering, transmitting, and 
deciphering an appropriate warning, and alerting the outpost forces. This 
was just what the shoguns intended. But just as a multitude of human 
errors on the part of Americans, cascading one atop the other, helped 
make tactical surprise perfect , so a series of similar human errors on the 
part of the Japanese deprived them of their last vestige of legality. 
Shortly after the attack commenced, Tadao Fuchikama, a messenger 
for the Honolulu office of R.C.A., picked up a batch of cables for delivery. 
He knew that the war had started and that it was the Japanese who were 
attacking the ships in the harbor, but he felt he had his job to do 
anyway. He glanced at the addresses on the envelopes, including the one 
marked "Commanding General," and planned an efficient route. Shafter 
was well down the list. His motorcycle progressed slowly through the 
jammed traffic; once he was stopped by National Guardsmen who had 
almost taken him for a paratrooper. At 11:45, almost two hours after the 
last attackers had vanished, Marshall's warning message was delivered 
to the signal officer. It got to the decoding officer at 2:40 that afternoon 
and to Short himself at 3. He took one look at it and threw it into the 
wastebasket, saying that it wasn't of the slightest interest. 
In Tokyo, Grew was awakened at 7 a.m. by the tele- 
phone , summoning him to a meeting at 7:30 with Togo. On Grew's 
arrival, the Foreign Minister gave him the Emperor's reply to the 
President. He thanked Grew for his cooperation and saw him off at the 
door. Four hours had elapsed since the attack had begun, but Togo 
never mentioned it. Shortly thereafter, Grew learned of the outbreak of 
hostilities from an extra of the Yomiuri Shimbun hawked outside his 
window. The Japanese soon closed the embassy gates . 
The Japanese in Washington destroyed their last machine and codes 
after encoding a final message that they were so doing—the last message 
sent on the Washington-Tokyo circuit, and read, of course, by the 
American codebreakers. But in Honolulu, police guarding the consulate 
after the attack smelled papers burning and saw smoke coming from 
behind a door. Fearing a conflagration, they broke in and found the 
consulate staff burning its remaining documents in a washtub on the 
floor. The police confiscated what proved to be the telegraph file plus five 
burlap sacks full of torn papers. These reached Rochef ort's unit that 
evening. Woodward was still working long hours in an attempt to break 
the PA-K2 messages that Mayfield had brought. Since the attack, the fear 
of sabotage had swelled to enormous proportions. "Nothing coming to 
light," his notes read, "so it was decided to reverse the process of 
deciphering, allowing for the encoding party to have either purposely 
encrypted the messages in this manner or possibly to have made an error 
in using the system employed due to confusion. This netted results." 
At about 2 a.m. on December 9, he cracked one of the messages 
picked up in the consulate. It was one sent from the Foreign Ministry to 
Kita on the 6th: "Please wire immediately re the latter part of my #123 
any movements of the fleet after the 4th." With this, he was soon able to 
unlock the other PA-K2 messages—including the long one setting up 
Kühn's light-signaling system. At about the same time in OP-20-oz, 
Kramer, who had been too busy with the 13 parts on Saturday to work 
on this message, was breaking out charts of Oahu and Maui to help in 
de-garbling the message, which was finally reduced to plaintext by 
Thursday. Marshall later said that it was the first message that clearly 
indicated an attack on Pearl to him— but this was, of course, after the 
fact.  
 
From:   Tokyo 
To:        Washington 
7 December 1941 
#902           Part 14 of 14 
 
(Note: In the forwarding instructions to the radio station 
handling this part, appeared the plain English phrase 
“VERY IMPORTANT”) 
 
7. Obviously it is the intention of the American 
Government to conspire with Great Britain and other countries 
to obstruct Japan’s efforts toward the establishment of peace 
through the creation of a New Order in East Asia, and especially 
to preserve Anglo-American rights and interests by keeping Japan 
and China at war.  This intentiona has been revealed clearly 
during the course of the present negotiations.  Thus the earnest 
hope of the Japanese Government to adjust Japanese-American 
relations and to preserve and promote the peace of the Pacific 
through cooperation with the American Government has finally 
been lost. 
 
The Japanese Government regrets to have to notify hereby 
the American Government that in view of the attitude of the 
American Government it cannot but consider that it is impossible 
to reach an agreement through further negotiations. 
 
JD-1:7143         SECRET     (M)  Navy trans. 7 Dec 1941 (S-TT) 
 
The fourteenth part of the Japanese ultimatum, as distributed to MAGIC recipients 
 
The information from it was immediately passed to counterintelligence 
units in Hawaii, where invasion was thought highly probable. Their 
agents interrogated residents in the neighborhood of the houses  
mentioned in the dispatch and listened to recordings of KGMB want ads, 
but found that the signal system had never been used. They arrested 
Kühn, who confirmed this. He was convicted on espionage charges and 
imprisoned at Leavenworth Penitentiary until after the war, when he was 
paroled to leave the country. 
On December 7, while Honolulu was still reeling from the devastation 
of the attack, F.C.C. monitors there picked up a Japanese-language news 
broadcast from station jzi in Japan. The announcer boasted of a " death -
defying raid" at Pearl, reported other events, and, about halfway through 
the broadcast, declared: " Allow me to especially make a weather forecast 
at this time: west wind clear."  
 
-11- 
 
and China at war. This intention has been revealed clearly 
during the course of the present negotiation. Thus, the 
earnest hope of the Japanese Government to adjust Japanese-
American relations and to preserve and promote the peace of 
the Pacific through cooperation with the American 
Government has finally been lost. 
The Japanese Government regrets to have to notify hereby 
the American Government that in view of the attitude of the 
American Government it cannot but consider that it is 
impossible to reach an agreement through further 
negotiations* 
     December 7, 1941. 
 
The last page of the Japanese note as typed by First Secretary Katzuso Okwnura and 
handed to Secretary of State Cordell Hull while Pearl Harbor was being attacked 
 
The O.N.I. translator noted that "as far as I can recollect, no such 
Weather forecast has ever been made before" and that "it may be some 
sort of code." It was the long-awaited winds code execute, apparently 
sent indicating war with Britain to make sure that some Japanese 
outpost that had not reported destroying its codes by the codeword 
HARUNA Would burn them. 
Shortly after noon in Washington on the day after the attack, the 
President of the United States stood before a stormily applauding joint  
session of Congress and opened a black looseleaf notebook. When the 
cheers had subsided into a hushed solemnity, he began to speak : 
 
Yesterday, December 7, 1941—a date which will live in infamy—
the United States of America was suddenly and deliberately 
attacked by naval and air forces of the Empire of Japan. 
 
He alluded to the fatal Japanese delay in delivering the ultimatum: 
 
The United States was at peace with that nation and, at the 
solicitation of Japan, was still in conversation with its Government 
and its Emperor looking toward the maintenance of peace in the 
Pacific. In deed, one hour after Japanese air squadrons hac 
commenced bombing in Oahu, the Japanese Ambas sador to the 
United States and his colleague delivered to the Secretary of State 
a formal reply to a recen American message. While this reply stated 
that i seemed useless to continue the existing diplomats 
negotiations, it contained no threat or hint of war 01 armed attack. 
 
The war was on. The most treacherous onslaught it history had 
succeeded. Japan had cloaked the strike fores in absolute secrecy. She 
had dissembled with diplomatic conversations and with jabs toward the 
south. She had—ir a precaution whose wisdom she but dimly realized— 
swathed her plans in a communications security so all enveloping that 
not a whisper of them ever floated ont But if the cryptanalysts had no chance to warn of th American lives before the war, they foum ample opportunities to exert 
their subtle and pervasivr talents during the struggle. In the 1,350 days 
of conflici in which an angry America turned Japan's tactical victory at 
Pearl Harbor into total strategic defeat, the cryptanalysts, in the words of 
the Joint Congressional Committee, "contributed enormously to the 
defeat of the enemy, greatly shortened the war, and saved many 
thousands of lives." 
That, however, is another story. 
 
3.  The First 3,000 Years 
 
ON A DAY  nearly 4,000 years ago, in a town called Menet Khufu 
bordering the thin ribbon of the Nile, a master scribe sketched out the 
hieroglyphs that told the story of his lord's life—and in so doing he 
opened the recorded history of cryptology. His was not a system of secret 
writing as the modern 
world knows it; he used no fully developed code of hieroglyphic 
symbol substitutions. His inscription, carved about 1900 B.C. into the 
living rock in the main chamber of the tomb of the nobleman 
Khnumhotep II, merely uses some unusual hieroglyphic symbols here 
and there in place of the more ordinary ones. Most occur in the last 20 
columns of the inscription's 222, in a section recording the monuments 
that Khnumhotep had erected in the service of the pharaoh Amenemhet 
II. The intention was not to make it hard to read the text. It was to 
impart a dignity and authority to it, perhaps in the same way that a 
government proclamation will spell out "In the year of Our Lord One 
thousand eight hundred and sixty three" instead of just writing "1863." 
The anonymous scribe may also have been demonstrating his knowledge 
for posterity. Thus the inscription was not secret writing, but it 
incorporated one of the essential elements of cryptography: a deliberate 
transformation of the writing. It is the oldest text known to do so. 
As Egyptian civilization waxed, as the writing developed and the 
tombs of the venerated dead multiplied, these transformations grew more 
complicated, more contrived, and more common. Eventually the scribes 
were replacing the usual hieroglyphic form