CHAPTER VII

GERMAN CODES AND CIPHERS

Secrecy is, of course, the most important consideration in the

German plots in this country. When Bernstorff wished to arrange

with Berlin to give Bolo Pasha ten million francs to betray his

country, he naturally did not write out his messages in plain

English for every wireless station on both sides of the Atlantic to

read them as they went through the air. He did, to be sure, write

the messages in English, and they looked plain enough--and innocent

enough--but they meant something very different from what they

seemed to mean. And when it got down to the actual transfer of the

money, another German agent in New York signed the messages, which

likewise were not what they seemed.

Those messages were in _code_. (They are reproduced and explained

in this chapter.)

Now _code_ should not be confused with _cipher_. When some Hindus

in New York, subsidized by Berlin, wished to write their plans

to some other Hindus in San Francisco, concerning their common

purpose of fomenting revolution against British rule in India, they

wrote out messages that consisted entirely of groups of Arabic

numerals.

Those messages were in _cipher_.

To any one but an expert, many code messages look simple and

harmless, and cipher messages usually look unintelligible and

suspicious. Yet, oddly enough, the cipher messages are by far the

easier to make out. Indeed, unless you have a copy of the code,

code messages can almost never be translated, whereas a straight

cipher message can almost invariably be unraveled by an expert,

if you give him enough time and material. Hence, by people who

know the subject (and nobody had mastered it so thoroughly as the

Germans), codes are used for secrecy, and ciphers are used simply

as an added precaution and to _delay_ the unraveling of a message

if, by any chance, the enemy has gotten possession of a copy of the

code.

German plot messages, therefore, are usually written out first in

plain German, then coded, and the code then put into cipher. Such

messages are called _enciphered code_.

For an enemy to get them to make sense, he has first to decipher

them, and then decode them. Any expert can decipher them--in time.

Decoding them is a very different matter.

Before taking up some of the German code and cipher messages that

have been translated, with dramatic results, it will be well to

discuss codes and ciphers in general.

A _code_ is an arrangement by which two people agree, when

exchanging messages, always to substitute certain words or symbols

for the real words of the message. Thus, they might agree on these

substitutions:

a = the

French ship = market

sailed from New York = price

sailed from Boston = quotation

to-day = is

for Marseilles = any even number

for Bordeaux = any number with a fraction

With such a code, a German spy in New York could cable a seemingly

harmless message to a friend in Holland, such as:

“The market price is 110.”

That would mean, of course:

“A French ship sailed from New York to-day for Marseilles.”

Whereas a very slight change in wording:

“The market quotation is 110¾.”

would mean:

“A French ship sailed from Boston to-day for Bordeaux.”

[Illustration: CODE MESSAGE TRANSMITTING MONEY TO SIR ROGER CASEMENT

In English it reads: “Embassy. 307-16, New York, April 10, 1916.

Mr. John Devoy has paid in $500 here with the request that they

be transmitted telegraphically to Sir Roger Casement. You are

respectfully requested to proceed accordingly and to charge the

amount to the Military Information Bureau. Receipt enclosed.”]

Messages of that sort could be exchanged daily between a broker

in Wall Street and a broker in Amsterdam, and, by the addition

of a few more words, could be infinitely varied and would look

like perfectly legitimate commercial correspondence. In fact, most

international business before the war (the Government now requires

that all messages appear in plain English) was carried on by coded

cables which turned long messages into short groups of words that

of themselves made gibberish. Several code books for business use

were on the market, containing hundreds of pages of these arbitrary

substitutions, which were useful, not for secrecy but for economy.

A dozen words could be made to say what normally would require five

hundred words.

Ciphers, however, have almost always been resorted to when secrecy

was desired. This sounds like a contradiction. But people who are

not experts use them because they think they are more secret,

since they look so. And experts use them when they are concerned

only with _temporary_ secrecy. They use them, then, because cipher

messages can be written and translated (by one’s correspondent)

without any equipment, like a code book, and much more rapidly than

code. Thus, if a general in the field wishes to send a message

ordering a colonel to advance in two hours, he sends it in cipher,

because it would take the enemy more than two hours to decipher the

message even if he intercepted it immediately, and because after

the two hours have elapsed the information in the message would be

of no value to him.

A _cipher_ is the substitution of some symbol for a letter of the

alphabet. The substituted symbol may be another letter--as writing

_e_ when you mean _a_. Or it may be a figure--as using 42 when you

mean _m_. Or it may be an arbitrary sign--as * to mean _c_. In

cipher, then, every word is spelled out, but the word _Washington_

might be spelled x=‖½?!^:°B if you had agreed that

w = x

a = =

s = ‖

h = ½

i = ?

n = !

g = ^

t = :

o = °

n = B

That is called a _substitution_ cipher, because some other letter

or symbol is arbitrarily substituted for every letter.

But another kind is called a _transposition_ cipher, because

in this the letters of the alphabet are simply transposed by

agreement--the simplest and most obvious example being to reverse

the alphabet, so that _z_ stands for _a_ and _y_ for _b_, etc. Such

a transposition cipher would read:

Alphabet of plain text abcdefghijklmnopqrstuvwxyz

Alphabet of cipher zyxwvutsrqponmlkjihgfedcba

and _Washington_ would be spelled _dzhsrmtglm_.

[Illustration: A letter from John Devoy, an Irish-American,

exposing his hand in a plot with the Germans to foment revolution

in Ireland]

Perhaps the cleverest transposition cipher ever devised--it is so

good that the British Army uses it in the field and, moreover,

has published text books about it--is the very simple “Playfair”

cipher. First a square is drawn, divided into fifths each way. This

arrangement gives twenty-five spaces, to contain the letters of the

alphabet--_I_ and _J_ being put in one square because there would

never be any plain sentence in which it would not be quite obvious

which one of them is needed to complete a word of which the other

letters are known.

[Illustration]

Next a “key word” is chosen--and herein lie the cleverness and

the simplicity of this cipher, because every time the key word is

changed, the whole pattern of the alphabet is changed. Suppose the

key word is _Gardenia_. It is now spelled out in the squares:

The second A is left out, as there must not, of course, be

duplicates on the keyboard. Now the rest of the alphabet is written

into the squares in their regular sequence:

[Illustration]

That is the complete keyboard. The method for using it is this:

The message is written out in plain text; for example:

DESTROY BRIDGE AT ONCE

(Only capital letters are commonly used in cipher work.) This

message is now divided into groups of two letters, in the same

order, so that it reads:

DE ST RO YB RI DG EA TO NC EX

(The X is added to complete the group and is called a _null_.)

These groups of twos are now ciphered from the keyboard into other

groups of twos, by the following method:

Where two joined letters of the original message appear in the same

_horizontal_ row on the keyboard, the next letter to the right is

substituted for each. Thus, the first two letters of our message

are DE. They occur in the same horizontal row on our keyboard.

Consequently, for D we write E, and for E we go “on around the

world” to the right, or back to the other end of the row, and write

G for E. This gives us DE enciphered as EG.

Where two joined letters of the original message appear in the

same _vertical_ row on the keyboard, the next letter below is

substituted for each.

Where two joined letters of the original message appear neither in

the same horizontal nor the same vertical row on the keyboard, we

imagine a rectangle with the two letters at the opposite corners,

and in each case substitute the letter found on the keyboard at the

other corner of the same horizontal row. This looks complicated,

but in reality is very simple. For example, take the third

two-letter group of our message--RO. The rectangle in this case is

R D E

B C F

L M O

and for R we substitute E, and for O we substitute L.

Substituting our whole message by this system, it reads:

Original DE ST RO YB RI DG EA TO NC EX

Cipher EG TU EL XC AB EA GR UM IF RZ

As telegraph operators are accustomed to send these gibberish

messages in groups of five letters (so that they can check errors,

knowing that when only four appear in a group, for example,

something has been left out) these enciphered groups of twos are

now combined into groups of fives, so that the finished cipher

reads:

EGTUE LXCAB EAGRU MIFRZ

The foregoing looks extremely complicated, but the truth is that

anybody, after half an hour’s practice, can put a message into

this kind of cipher (“Playfair” cipher) almost as fast as he can

print the straight English of it in capital letters. And unless

the person who reads it knows the key word which determined the

pattern on his keyboard, he would have to be an expert to decipher

it, and even he could do it only after a good deal of work.

Another ingenious cipher is called the “Chess Board.” First,

a sheet of paper is ruled into squares exactly like a chess

board--that is, a square divided into eighths each way. This

arrangement gives, of course, sixty-four small squares. Then, by

agreement between the people who intend to use this cipher, sixteen

of these squares are agreed upon and are cut out of the sheet with

a knife. Suppose, for example, this pattern is chosen:

[Illustration]

and the squares showing in white are cut out.

Next, another sheet of paper is ruled into a chess board, of

exactly the same size as the first. The perforated sheet is now

laid on top of the second sheet, so that the squares on the one

exactly cover the squares on the other. Now, with a pen or pencil,

the plain text of the secret message is printed on the under sheet

by writing through the perforations of the upper sheet, only one

letter being written in each square. This, of course, permits the

writing of sixteen letters of the message.

Suppose the complete message is to be:

“Authorize payment ten million dollars to buy copper for shipment

to Germany.” Then the lower sheet, after we have written through

the perforations, will look like this:

[Illustration]

The perforated sheet is now turned to the right through one fourth

of a complete revolution, so that the top of it is at the right

side of the lower sheet and so that the two chess boards again

“match up.” This operation exposes, through the perforations, a

new set of sixteen open squares on the lower sheet. The writing of

the message is continued, and the lower sheet now looks like this

(left):

[Illustration]

Again the perforated sheet is turned to the right, and sixteen

more letters are written. Once more, and the whole sixty-four

squares are utilized, looking like the last cut on the previous

page.

These letters are now put upright, like this.

[Illustration]

They are now read from left to right and from the first line down,

like ordinary reading matter. They are then grouped into fives for

telegraphic transmission, and an _X_ added at the end to make an

even five-group there. Thus the message, as transmitted, reads:

SADUL RRYAL TOHOF TRLNO IRNEI

MZNPI EEIPE PGOMC APYTU LAYHM

EBOOM NRNOT TESTX

When this message is received, it can, of course, be quickly

deciphered by printing it out on a chess board and placing over it

a sheet perforated according to the prearranged pattern.

This survey of codes and ciphers does not more than scratch the

surface of the subject, nor more than suggest the almost infinite

variations that are possible--in ciphers especially. It simply

gives a groundwork for an understanding of the German secret

messages now to be described.

[Illustration: EXTRACTS FROM A GERMAN CODE EXPERT’S BLOTTER

Showing the use of capital letters in the actual work of

enciphering a message, and the combined use of cipher and code]

Among the most interesting of these secret messages is the series

of wireless telegrams by means of which the German money was paid

to Bolo Pasha for the purchase of the Paris _Journal_--one of the

principal episodes in the treasonable intrigue for which Bolo was

recently executed by a French firing squad. These messages were in

English, and meant exactly what they said, except for the proper

names and the figures, which were _code_. To decode them, it was

necessary only to make the following substitutions:

William Foxley = Foreign Office

Charles Gledhill = Count Bernstorff

Fred Hooven = Guaranty Trust Company (New York)

$500 = $500,000

and to all other figures add three ciphers to arrive at the real

amount. For example, one of these messages read: “Paid Charles

Gledhill five hundred dollars through Fred Hooven.” This meant:

“Paid Count Bernstorff five hundred _thousand_ dollars through

Guaranty Trust Company.”

[Illustration: BOLO’S HANDWRITING

A letter written in New York to his bankers in transactions for the

purchase of the Paris _Journal_, with German money, the crime for

which he was shot]

The story of these messages is briefly this: Marie Paul Bolo

started life as a barber, became an adventurer and, in the

service of the Khedive of Egypt, received the title of Pasha for

a financial service which he rendered him. Returning to France

as Bolo Pasha, he married two wealthy women and lived in grand

style on their money. He became an intimate of Charles Humbert,

another adventurer, who achieved political power by questionable

methods and became a member of the French Senate. In the meantime,

the Khedive had been deposed by the British on account of his

pro-Turkish (and hence pro-German) activities after the Great War

began. Abbas Hilmi joined the colony of ex-rulers in Switzerland,

and there became a part of the German system of intrigue. He

received money from the Germans and, after he had deducted his

“squeeze” (which sometimes amounted to half the total), he paid

over the rest to Bolo, to be used by Bolo, Humbert, and ex-Premier

Caillaux in an effort to restore Caillaux to power and then to

further the propaganda for an early and hence inconclusive peace.

[Illustration: A TALE TOLD IN CABLEGRAMS

Code messages in the Bolo Pasha case, explained in the accompanying

pages]

[Illustration]

Either this method of supplying the French traitors with funds

became too dangerous, or the Germans preferred to keep their gold

and wished to use their credit in the United States to get American

gold for this purpose. In any event, Bolo Pasha appeared in

New York early in March, 1916. Strangely enough, this French

citizen bore letters of introduction to several Germans. The most

important was addressed to Adolf Pavenstedt, who was senior partner

in G. Amsinck & Company and for many years a chief paymaster of

the German spy system in this country. Through Pavenstedt, Bolo

met Hugo Schmidt, a director of the Deutsche Bank of Berlin, a

government institution, who had been sent to this country soon

after the war broke out to provide complete coöperation between the

older representatives of the Deutsche Bank here and the management

in Berlin.

Through Pavenstedt, as messenger, Bolo also got in touch with

Bernstorff, and arranged the details of the plan by which Bolo was

to receive 10 million francs from the German Government. He was

to use this money to buy the Paris _Journal_, which would then

be edited by Senator Humbert, who agreed to change its editorial

policy to favour an immediate peace. As the _Journal_ is one of the

most powerful dailies in France, with a circulation among more than

a million and a half readers, the sinister possibilities of this

scheme are readily seen.

[Illustration: MR. A. BRUCE BIELASKI

Who, as Chief of the Bureau of Investigation of the Department of

Justice, organized and managed the Government agents who unraveled

the German plots and captured the plotters]

Bernstorff committed the financial details to Hugo Schmidt. He, in

turn, asked Berlin by wireless for suitable credits in American

banking houses. These were arranged with the Guaranty Trust

Company and the National Park Bank--for many years American

correspondents of the Deutsche Bank. The amounts were then credited

to G. Amsinck & Company, of which Pavenstedt had long been senior

partner. He, in turn, placed them, with the New York branch of the

Royal Bank of Canada, to the account of Bolo Pasha. As the exchange

rate at the time ran in favour of American dollars and against

French francs, the 10 million francs (normally equal to about

2 million dollars) which Bolo got, required only $1,683,500 of

American money--which is just the sum of the amounts named in the

wireless messages.

The _Journal_ was actually bought by Bolo and Humbert, but before

they could do much damage with it, they were arrested, and Bolo has

already been executed.

[Illustration: THE COHALAN-IRISH REVOLUTION MESSAGE]

[Illustration: Above is the code message from Von Papen’s office

in New York to Bernstorff, transmitting a message from Justice

Cohalan, of the Supreme Court of New York, advising the Germans

upon the best means to make Sir Roger Casement’s revolution in

Ireland a success. On page 155 is the message written out and coded

for transmission. In English it reads as follows: “No. 335--16

_very secret_ New York, April 17, 1916. Judge Cohalan requests the

transmission the following remarks: ‘The Revolution in Ireland

can only be successful if supported from Germany. Otherwise,

England will be able to suppress it, even though it be only after

hard struggles. Therefore, help is necessary. This should consist

primarily of aërial attacks in England and a diversion of the

fleet simultaneously with Irish revolution. Then, if possible, a

landing of troops, arms, and ammunition in Ireland, and possibly

some officers from Zeppelins. This would enable the Irish ports to

be closed against England and the establishment of stations for

submarines on the Irish coast, and the cutting off of the supply

of food for England. The success of the revolution may therefore

decide the war.’ He asks that a telegram to this effect be sent to

Berlin. 5132 8167 0230 _To His Excellency Count von Bernstorff,

Imperial Ambassador, Washington, D. C._”]

The Hindus in this country, who were plotting with the Germans the

revolution that should destroy the British rule in India, used two

systems for their secret messages. The first was this _substitution

cipher_:

1 2 3 4 5 6 7

1 A B C D E F G

2 H I J K L M N

3 O P Q R S T U

4 V W X Y Z

The message, “Leave San Francisco” would be written, in this

cipher, as follows:

25 15 11 41 15 35 11 27 16 34 11 27 13 22 35 13 31

by giving each letter of the message the number to the left of it,

combined with the number above it.

The other system used by the Hindus was a _book code_. They agreed

upon a small English dictionary of a certain edition, and wrote

from it messages that were also groups of numbers, after this

fashion: 625-2-11 27-1-36 45-2-20 and so on. The first figure in

each group was the number of the page on which the word would be

found, the second figure gave the column, and the third figure was

the number of the word in the column, counting from the top of the

page.

But perhaps the most dramatic of all the intercepted messages

(except the Luxburg and Zimmerman notes, of which the story

cannot yet be told) were those which revealed the part played

by well-known Irish-American leaders in the ill-fated Casement

revolution in Ireland. The story of the Casement expedition is

too familiar to need to be retold. And comment upon the political

morals of Justice Cohalan and John Devoy becomes superfluous in the

light of these messages. American citizens (one of them signally

honoured with public office in New York), both held their Irish

blood superior, in their duty of loyalty, to the United States,

using their citizenship as a cloak under which to strike at Great

Britain, which has been for a quarter century the chief bulwark of

this country against Germany’s plan to conquer us and to impose

upon our country the most hateful tyranny in the history of the

world.

Fighting Germany's Spies · 第 7 / 11 章