RESPONSES to FLYNN'S cipher department are growing in number like the rolling snowball that doubles its weight at every turn. Readers are submitting their own ciphers as well as solving ours.
Unfortunately too many who submit them aren't ready to play fair with their audience. Their ciphers are too short. They are made arbitrarily hard.
But we are getting some worthy of passing on.
And we are taking your hints to make our own ciphers a bit harder to solve. Try this one for instance.
HE remarkably ingenious methods of secret writing described in this article are usually attributed to the Russian Nihilists, but as a matter of fact, their fundamental principle was the invention of the Greek historian, Polybius, who flourished in the second century B.C.
The several systems herewith described are all based on the same cipher alphabet, commonly known as the checkerboard cipher, no doubt because of its resemblance to the board used in the game of checkers.
The basic idea of this alphabet, that of arranging the letters in a square so that any desired letter may be indicated by other characters placed along two sides of the square, has been utilized by many cryptographers of various times. A cipher by Batista della Porta—1563—and the famous chiffre carre—cipher square— of Blaise de Vigenère—1586—may be cited.
And, besides, this ancient system contains the germ of many visual telegraphs, such as the semaphore and wig-wag, in military use throughout the world. It will thus be seen that we are deeply indebted to our old friend Polybius; and that in this one achievement alone, he has erected an ever enduring monument to his ingenuity.
The alphabetical key is given below, prepared, of course, for the English instead of the Greek alphabet. Otherwise it differs from the original only in unimportant details.
| SECOND FIGURE | ||||||
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | ||
| FIRST FIGURE | 1 | A | B | C | D | E |
| 2 | F | G | H | IJ | K | |
| 3 | L | M | N | O | P | |
| 4 | Q | R | S | T | U | |
| 5 | V | W | X | Y | Z | |
The person signaling the message according to the old Greek system, first held up on his left hand a number of torches, from 1 to 5, to show the row of any desired letter. Next he held up on his right hand in a similar manner a number of torches which would show the position of the desired letter in that row.
For example, P would be telegraphed by first holding up 3 torches on the left, followed by 5 on the right; L, by 3 on the left and I on the right; and so on. In sending a message, those torches not in use were concealed behind a wall of about six feet in height.
And in receiving, a sort of dioptical instrument consisting of two tubes was used, so adjusted that only the torches held up on the left could be seen through one tube, while those held up on the right would be visible through the other.
This ancient telegraphic system lends itself admirably for use as a cipher, in which the substitute for each letter consists of two figures: the first indicating the row, and the second the column of any desired letter, as in the subjoined example. You will note that, in the English key, both I and J are represented by the same number, 24.
| Message: | P | L | O | T | I | S | D | I | S | C | O | V | E | R | E | D | ||
| Cipher: | 35 | 31 | 34 | 44 | 24 | 43 | 14 | 24 | 43 | 13 | 34 | 51 | 15 | 42 | 15 | 14 |
It may seem a far cry from the battlefields of ancient Greece to the frozen wastes of modern Russia and Siberia, but the fact is that the Nihilists, in search of effective methods of secret communication, chanced to come across this particular system.
And the Russian mind, naturally subtle and alert, lost no time in working out many clever applications of the checkerboard key.
Among these were several telegraph systems not greatly differing from that of Polybius himself. Two prisoners in sight of each other could communicate by holding up the proper number of fingers in correct sequence.
Or at night a candle was used; its light being alternately obscured and exposed until the desired information had been conveyed in flashes from one prisoner to another.
Another important adaption of this system is known as the prison telegraph. By this method a prisoner could communicate with the inmate of an adjacent cell by tapping on the wall between them. The word Plot (35-31-34-44), for example, would be telegraphed thus: (P) three taps, pause, five taps, longer pause; (L) three taps, pause, one tap, longer pause; and so on.
This system came to be used outside of Russia, and is said to be still practiced in this country. Even a whispered conversation in cells is forbidden by the American penal system, thus necessitating some secret mode of communication.
At night a prison becomes a place of weird and subdued sounds, intermingled with stealthy rappings and tappings, which are, of course, heard by the watchful keepers, but cannot be accurately located. In this manner plans for a break, or plots for crimes to be committed when the prisoners again are free, are relayed from cell to cell.
But to return to Russia, the Nihilists went even further with their checkerboard key, and devised a cipher of masterly ingenuity of the double key variety. Ciphers of this type are so called because a message, after being enciphered by a first alphabetic key, is further obscured by an additional operation controlled by a secondary key, which usually consists of a secret key word known only to the correspondents.
So confident were the Nihilists that this cipher was absolutely unbreakable, that they intrusted their gravest messages to it without the slightest fear that they would ever be read without the secret key word, even if they fell into the hands of the officials of the Russian Cipher Bureau.
The complete method of operating this cipher is shown below.
In line (1) is written the message to be enciphered, "Plot Is Discovered."
At (2) are entered the numerical substitutes for these letters, as found in the checkerboard key by the process already described.
At (3) the letters of the secret key word decided upon by the communicating parties are written under the letters of the message, being repeated as many times as the length of the message (i) requires.
At (4) the numerical equivalents of the letters of the secret key word are entered, being found in the checkerboard key in exactly the same manner as were the numbers of the original message in line (2).
Finally at (5) the numbers in lines (2) and (4) are added, giving the numbers of the final cipher: 67-65-77-57-58, et cetera.
| (1) | P | L | O | T | I | S | D | I | S | C | O | V | E | R | E | D |
| (2) | 35 | 31 | 34 | 44 | 24 | 43 | 14 | 24 | 43 | 13 | 34 | 51 | 15 | 42 | 15 | 14 |
| (3) | M | o | s | c | o | w | M | o | s | c | o | w | M | o | s | c |
| (4) | 32 | 34 | 43 | 13 | 34 | 52 | 32 | 34 | 43 | 13 | 34 | 52 | 32 | 34 | 43 | 13 |
| (5) | 67 | 65 | 77 | 57 | 58 | 95 | 46 | 58 | 86 | 26 | 68 | 103 | 47 | 76 | 58 | 27 |
In practice the letters of the key word in line (3) may, of course, be omitted. And in deciphering, the process as above described is merely reversed. The recipient writes under the numbers of the message the numbers of the secret key word.
He then subtracts the latter from the former, obtaining the simple numerical substitutes which can then be deciphered directly by the checkerboard key.
The Nihilist cipher in its more elementary forms is a simple substitution cipher. That is, it is a cipher in which a given letter of the alphabet is always represented by the same cipher character; and in which a given cipher character always represents the same letter of the alphabet. But the double key variety last described, is a poly alphabetical substitution cipher.
In other words, it is a cipher in which the concealment of the message is effected not by one, but by several different alphabets used in succession.
In a cipher of this type a given letter of the alphabet may be represented by several different cipher substitutes; and conversely, a given cipher substitute may stand for several different letters of the alphabet. In the above short illustrative specimen, for example, the letter O is represented by both 77 and 68, and S by 95 and 86; while 58 is the substitute for both I and E.
The cryptogram given below for solution has been enciphered by the Nihilist variable key cipher exactly as described in this article, with the exception that another key word than "Moscow" has been used. To expect you to decipher this message without the secret key would perhaps be asking too much. But to hand you the key outright would spoil the fun. Therefore, as a sort of compromise, we will hide the key under the door mat by causing you a little study in figuring it out.
Here is the key in Nihilist variable key cipher, enciphered by means of itself as a key:
42-62-108-66-66-86
If you think this leaves you out in the dark as much as ever, as an experiment just try enciphering any word you please by means of itself, and carefully study the results.
The cryptogram, a personal message to every cipher fan, follows:
45-52-108-67-78-95-45-62-85-76-48-76-35- 42-67-75-87-78-65-65-76-75-44-75-73-73-78- 77-77-58-54-55-87-77-56-58-54-55-77-57-64- 67-64-75-67-57-68-66-36-73-98-67-77-66-36- 46-68-57-77-77-63-65-75-77-56-67-64-45-69- 68-44-85-65-63-69-66-77-66-36-83-78-64-64- 58-54-45-69-44-84-77-63-75-88-47-48-56-45- 66-77-48-75-67-65-52-88-75-87-77-66-83-78- 77-56-77-66-75-98-56-48-68-36-85-106-67- 75-57
If you succeed in deciphering the above brain sprainer, send in your solution to the editor of this department. All those received in time will be printed with the next article.
The correct solution of the Nihilist cipher, and a new article dealing with one of the most famous ciphers ever invented, will appear in an early issue. Watch for further announcement.
The cipher in our last article was one of cipher by that letter next following it in the simple substitution type, in which each the alphabet. The entire alphabet was thus letter of the message was represented in shifted forward one place, B being used for A, C for B, and so on until the message had been worked out. The only exception to this rule in the original Augustus cipher was that he used AA as the substitute for Z, as shown in his complete cipher alphabet subjoined:
| Message: | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z |
| Cipher: | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | AA |
The most frequent cipher letter as shown by the frequency table was F, which occurred fourteen times. Assuming that this was the substitute for E, the letter next preceding it in alphabetical order, and apply ing the same principle to the other letters of the cipher, the translation, as given here with, follows without any difficulty:
"THERE NEVER WAS A TIME WHEN MEN COULD AFFORD TO TELL THEIR SECRETS TO THEIR ENEMIES."
Correct solutions to the Augustus Cipher in FLYNN'S for February 21 were sent in by the following readers in time for publication in this issue:
We are in receipt of an interesting letter from Mr. Frank W. French, concerning a cipher of his own invention, which is appended in full:
DEAR SIR: Niles, Mich.
The following may be used if you wish—for publication with the understanding that I am not to explain how it is done until after patents are secured:
| Message: | J 5 Z 1 Z J V H B P Z Z P P F F M Z 2 L J 5—R L |
| Key: | The word "Eureka." |
| Translation: | Repetition of the word " Eureka." |
We are so sure it is "unpickable" that we here give message, key, and translation, and defy any one to tell how to use the key to get the translation. The system can he written or translated at the rate of from sixty to one hundred letters or figures per minute, by the ones for whom intended only.
Yours truly, FRANK W. FRENCH.
In commenting on Mr. French's cipher, it must be mentioned that, considered from the angle of utility and not of structure, ciphers fall into two general classes.
To the first class belong those ciphers which consist of a more or less simple key, knowledge of which by any unauthorized person would frustrate the secrecy of the cipher, and thus render it of no further value.
The Augustus Cipher described in FLYNN'S for February 21, the solution of which is in this issue, is one of this type. It is easy to see how the field of usefulness of such ciphers is necessarily limited.
The second class includes those ciphers which are capable of many variations by means of a secondary key that can be changed at will, and possession of which is required in order to read the cipher in the usual manner. The Nihilist variable key cipher in this issue is such a cipher.
All ciphers intended for public use must obviously be of this second class, for the reason that it is possible for any interested person to obtain a copy of the cipher system, and become familiar with its use. To illustrate just what is expected of such ciphers, let us consider army ciphers, which of all others are expected to stand up satisfactorily under the most rigorous conditions.
First, it is taken for granted that the enemy is familiar with the cipher, either by capture during an engagement, through their espionage system, or otherwise.
It is further considered that it is possible for them to intercept five or six hundred messages per day during actual field operations.
All they lack is the secret key word. Under such conditions it is generally accepted as true that any cipher of this kind can be solved if sufficient time is available.
From this it will be seen that even the best of such ciphers are not expected to be absolutely indecipherable. All that is required is that any solutions will be so delayed by difficulties encountered in deciphering, that any information discovered from the messages will no longer be of value for the reason that such events as may have been mentioned will have already taken place.
That a single message in such a cipher should resist all attempts at solution is admitted without any argument. But this does not mean that the cipher as a system is unbreakable. It only means that when the material available in any one key is limited, occasional messages may be indecipherable.
You speak of having your cipher patented. It may be well to mention here that no device can be patented that does not involve some mechanical feature. If your cipher can be arranged in the form of a cipher disk, a slide rule, or other mechanical device, it is patentable.
But if a printed description is all that your cipher will require, the Patent Office will only allow a copyright.
Finally, in as much as it is often desirable to send messages by wire, any cipher intended for general purposes should make use of only such characters as are found in the telegraph alphabets.
Also, in all ciphers intended for telegraphic transmission it is necessary to arrange the cipher characters in groups of five, as this is the basis upon which telegraph companies base their charge, each group of five characters being counted as one cipher word.
The following letter is so good that we must pass it on to our readers:
Toronto, Ont., Canada.
DEAR SIR:
I was keenly interested in the first of your new series of articles in FLYNN'S, "Solving Cipher Secrets." I never knew until I read your article that systems of secret writing were used as far back as the days of ancient Sparta. In fact, I should have said that ciphers were the invention of comparatively recent years.
In my opinion FLYNN'S just needed a series of articles such as yours to "put it over." The detective fiction is first rate, but the cipher secrets add that little touch of variety which is necessary to a top notch magazine.
Hoping to see many more of your delightful articles in future issues, I am
Yours truly,
C. H. HOWARD.
Many thanks, Mr. Howard, for your kind words of appreciation. Just such letters at yours will induce us to expend still greater efforts to please our ever-increasing circle of cipher fans. Tell your friends about the new department, and be sure to watch for our next article; it will be of special interest, Write us often.