|
CODES AND CIPHERS
ALEXANDER d'Agapeyeff
(1902-1955)
Compass Book No, 1
Geoffrey Cumberlege
Oxford University Press
1949
I want to thank Geoffry and Marjory Cass who not only encouraged me by showing interest in my writing, but who also found some material which proved very valueable.And especially I want to express my gratitude to Rachel Wood, without whose help in research and constant co-operation throughout, this book would never have been written.
Alexander d'Agapeyeff
CONTENTS
Chapter
I. The Beginning of Cryptography 7
II. From the Middle Ages onwards 22
III. Signals, signs and secret languages 48
IV. Commercial Codes 70
V. Military codes and ciphers 89
VI. Types of codes and ciphers 103
VII. Methods of deciphering 121
Bibliography 145
Index 147
I. THE BEGINNINGS OF CRYPTOGRAPHY
7
'A man is born without any languages and yet is capable of all.'
Bishop Wilkins
Codes and ciphers appear at first sight to be such complicated and difficult affairs, and so completely wrapped in mystery, that all but the boldest hesitate to tackle them; yet as Bishop Wilkins implied in the sentence quoted above, any person of average ability is capable of acquiring a knowledge of almost any subject, and there is no reason why we should not, with the exercise of patience and perseverance, become familiar with the main principles of Cryptography - to give this fascinating science its proper name.
Most people have a natural curiosity to know what lies beyond the closed door, what secrets are hidden behind signs and symbols that have no obvious meaning. When the urge to solve such problems becomes a dominating force in a man’s life, then he may discover new worlds. It may be that there is nothing much left to [be] discover[ed] in the material world, but there are unlimited possibilities in the world of thought[s] and idea[s].*
*[..] differs from the 1939 edition.
But thought, if it is to be effective, must be controlled and disciplined, and a knowledge of codes and ciphers can only be acquired by means of orderly and patient thinking. The same habit of thought is of incalculable value in many walks of life, so what is taken up as a hobby and amusement may be a training for more seious things.
The history of cryptography is very old - almost as old as writing itself. In the library of the British Museum and elsewhere there are books on the subject written hundreds of
[A]
8
years ago by men whose names are forgotten, but whose orderly precision of thought provides us with a basis for practically every cipher we now use.
Cryptography itself dates back to the oldes scriptures and fables that have come down to us, and it was closely related to the earliest attempts at writing.
'The noblest acquisition of mankind is speech,' wrote Astles in the last century, 'and the most useful is writing. The first eminently distuinguishes man from the brute creation, and the second from uncivilesd savages.'
The oldest scripture of India, the Vedas, claims that Brahma himselff first communicated letters to mankind. We find also that the great Greek philosopher Plato attributed the invention of letters to the god Zeus (Theuth), while another assumed that Hermes (called by the Egyptians Thoth) taught letters to man.
The first mention of letters in our Bible occurs in Exodus XVII. 14: 'And the Lord said unto Moses, Write this for a memorial in a book...' This passage gave rise to a discussian whether it was God who gave letters to Moses, or whether Moses already knew how to write.
However, all these legends show very clearly the value that mankind placed on the art of letters. Otherwise it is unlikely that these different scriptures of various races, separated from each other both in time and in space, would have attributed, even indirectly, the invention of writing to divine powers.
In the olden days secret writing was not so necessary as it is to-day, as very few people could read or write, and doubtless at first it was only a few priests and scholars who used it.
We know that even in the early days of Egypt there already existed, apart from hieroglyphs, two kinde of writing; the hieratic or sacred writing, used secretly by the
9
priests, and the demotic or vulgar writing, used by the people. The former writing was at all times guarded from the people, and it is recorded that one of the Pharaohs found it necessary to issue an order forbidding the priests to teach the secret letters to the ordinary citizen.
Plutarch mentions a ‘famous inscription’ on an obelisk outside a temple at Sais, dedicated, he says, to Minerva; on this were painted an infant, an old man, a hawk, a fish, and a ‘sea-horse’ which was the name given to hippos by the Ancient Greeks. The meaning of these figures, he tought, was that young and old know that God hates impudence. An eighteenth-century author translated this differently. ‘The infant, which is the first figure, represents man’s first entrance into the world, and yje old man implies the going out of it; the falcon represents God; the fish, harred, because of its association with the hated sea, which symbolised storms; and the sea-horse, murder, violence and injustice, because by ancient fables the sea-horse murdered its own sirewhen it became of age.’ The meaning he says may be: ‘O you that enter the world and go out of it, know that God hates injustice !’
Apropos of this story, I have found in Pietri Valeriano’s 1575 edition of Hieroglyphica a drawing supposedly illustrating the obelisk described by Plutarch. It is reproduced here [afb] The fact that Plutarch calls it an obelisk may indicate that there was some traditional Egyptian story connected with it, and perhaps he based his version on an eyewitness’s account.
10
However that may be, the details on the obelisk are quite obviously pure imagination. Plutarch says himself that it was found on a temple at Sais, but later authors, copying this passage from his works, for some reason mention Thebes. Sais, however, was under the Nile mud in Plutarch's time, so that he could never have seen the Original inscription. At Thebes nothing of the kind has ever been found.
Plutarch himself did not know how to read hieroglyphs; the conventionalized part of them would have no meaning for him, for by his time all knowledge of hieroglyphs was already lost. And if some traveller had ever seen or copied some kind of an inscription Plutarch would recognize only a part of it. In such an hieroglyph as this:
[Afb.]
(meaning 'upon the') he would recognize only the child's head, and in this:
[Afb.]
(meaning 'head' or 'god') he would understand only that it was an old man because of the beard.
[Afb.]
11
In this inscription (-envelop) the only figure he would recognize would be the fish; while in this:
[Afb.]
(which means ‘a moment of time’) he would see only the hawk and the hippo’s head. We now see where Valeriano got his head of a hippopotamus, which led me to try to reconstruct this.
Any attempt to do so, however, is so much of a speculation that, not being an Egyptologist, I had to give it up. But it is quite safe to assume that the inscriptions (if any) mentioned by Plutarch certainly did not have the meaning ascribed to it. It is also quite clearthat the authors who copied Plutarch never tried to verify their facts.
In brief, Plutarch was writing in Ignorance.
Symbolism has always been of great influence among Eastern peoples. The sending of a bird, a mouse, a frog and an arrow by the Scythians to the Persians was a gentle hint to them that unless they could escape as birds, could swim as frogs, or conceal themselves as mice, they were hastening to swift destruction.
The necessity of sending messages in a form which could be understood only by a select few was apparently realized very early in history, and it was from this necessity that arose the science of cryptography.
Cryptography (from the Greek kryptos, secret, and graphe, writing) is a manner of conveying messages in a form which will prevent anybody reading them except those who have the kkey.
The form used may be a code, i.e., a set of letters or words with arbitrary agreed-upon meanings for brevity or secrecy;
12
or a cipher (Hebreww saphar, to number) meaning secret writing, either in invented characters, or characters given different powers, or even figures with agreed meanings.
Thu, if we agree that we shall use 'Tom' for 'Arriving by the 12 o'clock train tomorrow', ''Jack' for 'Leaving here by car at 4 p.m. today', this will be a code; while if we agree that the figures 1, 2, 3, 4, ... stand for a, b, c, d, ... etc., then 25,5,19 will mean 'yes' and 14, 15 will mean 'no', and thus will be a cipher.
There are naturally many ways of 'coding' or 'enciphering' a message. Although their employment by armies for military purposes excites the imagination, secret ciphers were used much more in the olden days for teaching secret knowledge, and in later times for political and diplomatic purposes. In modern times they are employed for purposes of trade and commerce.
In this connexion it must be realised that codes are not used in business only for the sake of saving money on telegrams, but also to keep certain information from competitors. In consequence a great many private codes are used.
We find, however, that secret ways of sending messages for peaceful purposes were known at the beginning of civilisation. In the Bible both Isaiah and Jeremiah used a form of cipher to hide the real meaning of their prophecies from the Babylonians.
In the Middle Ages a very interesting study was made of what is known as the 'hidden'' sacred mysteries of the Scriptures'. This study, known as cabbalis, was at that time regarded as almost a science. For instance, in Genesis XLIX, 10: 'The sceptre shall not depart from Judah, nor a lawgiver from between his feet, until Shiloh come; and unto him shall the gathering of the people be', the initial letters
13
of unto him shall the gathering of the people be’’ spell Hebrew thw word ‘Jesus’.
In the story of the Creation in Genesis I, in the same manner the word ‘truth’ is spelled six times. Also a very ‘important cabbalism occurs in Exodus III, 13: ‘And Moses sait unto God, Behold, when I com unto the children of Israel, and shall say unto them, The God of your fathers hath sent me unto you: and they shall say to me, What is his name ? What shall I say unto them ? The final letters of the words in the last line of this verse form in Hebrew the word ‘Jehova’, which was God’s secret name. Julius Caesar used this kind of cipher in his military correspondence, writing ‘d’ for ‘a’, ‘e’ for ‘b’’, etc., and called that quartam elementorum literam.
We can be fairlu confident in stating that signals for sending messages at a distance were used as long ago as 1184 B.C., and this was quite apart from the legendary use of such signals by Medea in het conspiracy against Jason. Actually the reading of the Trpjan war episode makes one believe that some sort of flash or mirror apparatus was used to give the signal to unlock the wooden horse of Troy.
Again Agamemnon informed Queen Clytemnestra by fire signals that Troy was taken. This is probably the first example of an agreed code.
Some time later the Spartans during their wars, used a form of staff called ‘skytale’, round which a roll of parchment was wound slantwise. Letters written seemingly at random on the edges of the unrolled parchment could be consecutively read forming complete words when the roll was wound again round another staff of equal diameter, the thickness of the latter giving the necessary distance for the appropriate letters to come together. To us this does not appear a very secret method of writing, ast it can easily
14
deciphering by juggling together the edges of the roll; but doubtless at the time of Alcibiades and Lysander, when the invention was new, knowledge of reading was sufficiently rare to ensure the secrecy of the 'skytale' method.
[Afb.]
In those early days many strange and often cruel methods of sending secret messages were employed. A certain Histiaeus, Greek ambassador in Persia,wishing to send secret political information to his country, shaved a slave's head, and under pretext of curing his bad eyes, branded a message on his skull, let the hair grow again, and sent him to Greece, telling him that when Aristagoras shaved his head the second time his eyes would recover. Others used to drug their slaves and brand messages on their backs, and when they came to they wre sent to their destination, having been told that when thet arrived and uncovered their backs a healing ointment would be given to them. In both cases the slaves knew nothing about the important messages they were carrying.
Tacitus describes several ways of sending messages into a besieged city, such as using a manuscript instead of a bandage on a wound, sewing a letter in the soles of a person's shoes, rolling thin leaves of lead bearing the written message into ear-rings, and writing on wooden tablets which were then covered with wax. All these, however, were not so much examples of secret writing as of writing carried secretly.
About 350 B.C. the first telegraphic machine was improvised by Aeneas Tacitus. It consisted of a narrow earthen vessel filled with water, which could either be added to or
15
drawn off at will. A piece of stick was thrust through a cork floating on top of the water, and at the end of the stick was a torch. This stick was marked with notches about three inches apart, and each notch stood for some common event that happens in war.
If water was drawn off from the vessel, the stick would descend by so many divisions, and an observer, noting the distance the stick had travelled, could tell which of the expected events had occurred.
Later Polybius improved on this method by a system of torches together with a special code consisting of five groups of five letters each:
[Afb.]
The signaller was provided with ten torches, five for the left hand and five for the right; which the latter he signalled the number of the group, and with the former the letter in this particular group, so that two torches in the right hand meant group No. 2, and four torches in the left hand meant the fourth letter in that group, which is I.
The great advantage of this was that any messages could be sent, and not merely messages that had been pre-arranged, as in the case of the notched stick.
It may surprise some of us to realize how very early this kind of telegraphy was invented; but we find also that at an even earlier date propaganda, which we regard as belonging
16
exclusively to the twentieth century, was used by a Spartan king at the siege of Trezene. He ordered his soldiers to shoot blunted arrows into the town bearing this inscription: 'I come to liberate you.' Reading these signs on the arrows, the discontented inhabitants revolted against their tyrant, and eventually opened their city gates to the Spartans.
A few hundred years later, when the Roman Empire was at its height, a great advance in cryptic writing was made by a freed slave of Cicero's, Tullius Tyro, whose name was immortalized through his invention of the first method of shorthand, in which he used arbitrary marks for whole words. Although some historians say that he only perfected the method which was already known to the poet Ennius, nevertheless the characters used in Germany as late as the tenth century were called Tyronian after him.
In this connexion a curious story is told concerning the discovery of an old psalter inscribed with Tyronian characters, but labelled at the time of Pope Julius II 'Psalterium in lingua Armanica' by some monk who mistook Tyronian characters for Armenian writing.
[Afb.]
We give a few of these characters, which were non-alphabetical but so well systematized that words beginning or ending similarly had the same kind of marks.
17
Probably this may have given rise amongst the Romans to the excessive use of abbreviations of words, such as they inscribed on statues, coins, and so on; but this was obviously done not so much for reason of secrecy as to put the greatest amount of information in the least amount of space. This tradition is still carried on in the present day on our coins: DEI GRA: BRITT: OMN: REX: HD: DEF: meaning 'Dei Gratia Britanniae Omnis Rex Fidei Defensor'.
When the art of painting ikons was flourishing in Byzantium, an abbreviated form of writing the names of the saints to whom the ikons were dedicated was employed. In Slavonic Russian, as in Latin, a wavy sign was used over the abbreviated word, thus:
[Afb.]
Doubtless one can trace the origin of these abbreviations to the time when early Christians were persecuted in Rome, and used to hide their votive images from the pagans under incomprehensible marks known only among themselves. Quite possibly this may have been one of the origins of the Orthodox and Roman Catholic custom of making the sign of the cross, which was used as a secret sign to show their adherence to the Christian faith.
It may be that the Fascist salutes, the Communist clenched fist, the Chinese bow and the Mohammedan salaam also belong to the same category.
Another example of the ingenuity of the Romans, at the time of the Roman conquest of Britain, was the building of Hadrian's wall to keep out the savage Picts. It was over a hundred miles long, and watchtowers were placed along it about a mile apart. Inside the thickness of this wall long hollow pipes and trunks were built through which the sentries passed signals to each other by tapping on the hollow
[B]
18
tubes and using Polybius' code, or even just shouting into them, the pipes carrying the sounds along to the next tower.
Walchius, a writer in the Middle Ages, pondering on this system of Roman telephony, came to the astonishing conclusion that if words were spoken into a hollow tube which was sealed at one end and then corkef at the other, the sound might remain in the tube until it was uncorked again, when the words would flow out in their proper sequence. As far as we know, this notion has never been put into practice. Nevertheless in this story we seem to get a hint of the modern gramophone record.
[Afb.]
Another set of secret signs which exercised the curiosity not only of ancient writers but also of some moderns were the so-called mason-marks, found on medieaval cathedrals and other contemporary buildings; surprisingly similar marks were found also on the old Egyptian pyramids and temples. We reproduce a few of the here, and although noody really knows what they mean, it has been suffested that they were the marks of actual individual masons who were paid at piece-work rates. As can be seen, the only rule about them was that they had to be made up of at least one angle. Agreat many people read into these signs some occult significance or associate them with freemasonry.
Perhaps we may regard trade-marks and hall-marks on silver as the present-day descendants and sole survivors of these ancien mason-marks.
An alphabet called the 'Oghama', based on straight and
19
inclined vertical lines, over, under, or through a horizontal line, is supposed to have been used in the pre-Christian era by the Celts; and many stone monuments can be seen in Ireland and Wales bearing these curious incisions. Nobody has hitherto bothered very much anbout these ‘Oghams’, but we reproduce a set of them as an example of an invented alphabet which was actually used many centuries ago.
[Afb.]
In the fifth century the fabulous Pharamond, reputedly 43rd in succession to the even more legendary Marcovir, king of the Franks, was credited with the invention of several secret alphabets. Actually a few of the were used by the great Charlemagne when communicating with the vassal kings of his vast empire. Some of the letters are preserved to this day in Paris. They seem rather uncouth, and as they certainly required a good deal of time to encipher, they were easily misread and were soon given up. The older transposition ciphers, which were used by the Syracusans and Carthaginians, came into general use again for diplomatic and political purposes. Though they were simple, the decline of learning in the dark ages made it less likely that any unauthorized person would try to decipher them.
We come again to the Bible, on which by tradition the first transposition ciphers are based. It is here that the cabbalists read the word ‘Sheshach’ for ‘Babel’ (Jeremiah XXV. 26). ‘Babel’ is written with the second and twelfth letters of the Hebrew alphabet, whilst ‘Sheshach’ is written with the second and twelfth letters from the end. Isaiah concealed his prophecy from the people (Isaiah VII. 6.) By
20
writing ‘Tabeal’ for (King) ‘Remaliah. This method is called ‘Athbash’, and the word itself is the key to the kind of transposition used, for ‘A’ is the first letter of the Hebrew alphabet and ‘TH’ is the last while ‘B’ is the second and ‘SH’ is the second from the end, etc. There is another kindwhich is known as ‘Albam’, where the middle letter ‘L’ is used instead of the first letter ‘A, and the letter ‘M’ is used for ‘B’, etc. Such simple transposition ciphers were used up to the seventeenth century.
21
of water can be relied upon to make an invisible ink. A solution of alum would also have the same result. Both these inks clearly appear on damping the paper, and the aqua fortis ink disappears again as the paper dries.
In case of emergency, these chemicals may be difficult to get hold of, but there are ordinary household articles that also make excellent invisible inks. The juice of a lemon, of an onion, vinegar or ammonia, used as inks, all remain invisible until they are heated. In the case of the onion juice, however, not only is it uncomfortable to prepare, but if a plain piece of paper recking of onion fell into the hands of the most unsuspecting person it might easily give the secret away.
Another obsolete method is to incise letters on some soft wood-poplar, for instance- to the depth of a quarter of an inch, and then immediately afterwards either beat the wood with a hammer until the letters can no longer be seen or put it in a vice. To read the message the wood must be damped, and the incised letters will appear again.
22
II. FROM THE MIDDLE AGES ONWARDS
Towards the end of the fourteenth century Europe was reorganizing itself. The nations started to assume some of theforms we know to-day. There were wars and excursions and a great deal of diplomatic activity, as civilization and learning began to emerge from the chaos of pre-Renaissance. For these reasons the old transposition ciphers, as well as the clumsy methods of invented alphabets, were no longer of muche use; they were too easily deciphered and took too long to deal with. There was a need of a more scientific approach to this subject, and out of the art of cryptography man had to evolve the science of codes and ciphers.
At that time all science was more of a mystery than it is to-day. Man had just begun a rational search into the secrets of nature; but thatsearch seemed very close to magic, and it was only the very few enlighted and fearless people who dared to look into the unknown. It is due to the early alchemists that we have our modern Chemistry, and to astrologers that we have the science of astronomy.
The first book dealing with cryptology was written in 1499 by John Tritheim, Abbot of Spanheim, a small town in Germany. There is little doubt that Tritheim merely collected the ciphers already used by certain European courts, and as he was familiar with occult matters, he also proposed a kind of first code based on cabbalistic words wherein he tried to hide the real meaning under cover of a mysterious language.
23
The courts became frightened lest Tritheim in his Polygraphia - or as he called it 'a manner of occult writing' - was giving away too much, so they started a campaign of abuse against him, stressing his magic practices; and the leaders of several religious bodies persuaded Frederic II of the Palatinate, who was also interested in the subjects and had the Original manuscript in his library, to burn it publicly.
This was done with great pomp and ceremony, and popular feeling was roused to such an extent that it was very Lucky for the Abbot that he escaped being burnt as well.
However, the first edition in Latin was printed in 1518, and a French translation was published as early as 1561, by Collange; it was very soon followed by a German translation.
Actually Tritheim, being a Benedictine Abbot, only carried out a Benedictine tradition. This order, in the ninth century, published a Treatise of Diplomacy in which they described one of their own ciphers based on replacements of all the vowels by dots, 'i' by one dot, 'a' by two, 'e' by three, 'o' by four, and 'u' by five, so that 'Bonifacia' would read 'B::n.f:c:'. This idea is obviously not very complicated, and it rarher suggests that the Benedictine monks had other ciphers which they preferred not to publish.
The first part of Tritheim's Polygraphia consisted of a number of code words for each letter of the alphabet, but arranged in such a manner that if each letter of the message was replaced by a code word, the result was a complete sentence having an innocent meaning.
Below will be found fourteen coded alphabets illustrating the way they were meant to be used:
24
TRITHEIM'S CODE ALPHABETS 1st 2nd 3rd 4th
A Jesus Immortal Producing Angels
B God Omnipotent Saving Archangels
C Saviour Compassionate Illuminating Saints
D King Ineffable Conferring Spheres
E Pastor Universal Moderating Heavens
F Author Almighty Expressing Sea
G Redempter Magnificent Governing Earth
H Prince Puissant Disposing(of) World
IJ Maker Just Dominating Men
K Conservator Sempiterneal Creating Sun
L Governor Celestial Cognising Moon
M Emperor Divine Guiding All
N Moderator Excellent Blessing Hierarchies
O Rector Triumphant Constituting Bodies
P Judge Clement Confirming Spirits
Q Illustrator Peaceful Conducting Souls
R Illuminator Pacific Sanctifying Times
S Consolator Invisible Honouring Humanity
T Sire Eternal Ministrating Ages
UVW Dominator Invincible Exorcising Eternity
X Creator Benign Elevating Firmaments
Y Psalmist Pitiable Sustaining Stars
Z Sovereign Incomprehensible Vilifying Air
& Protector Excellent Ordering Cosmos
5th 6th 7th 8th
A Gives (Tothe) Christians Eternal Life
B Delivers Requiring (needy) Perpetual Joy
C Attributes Faithful Infinite Joyousness
D Increases Attendants Angelic Glory
E Presents Righteous Immortal Consolation
F Renders Penitents Enduring Felicity
G Remits Good Incomprehensible Beatitude
H Renders Supplicants Incorruptible Jubilation
IJ Envoys Hopeful Durable Tranquility
K Transmits Patient Permanent Amenity
L Administers Afflicted Ineffable Recreation
M Permits All Celestial Clarity
N Inspires Tormented Divine Union
O Retributes Perturbed Interminable Peace
P Orders Desolated Perfect Light
Q Contributes Mortals Sincere Glorification
R Frees Humans Pure Benediction
25
S Confers Languishing Glorious Security
T Manifests Repentant Supernatural Favours
UVW Reveals Catholics Indicible Fruition
X Maintains In the World Peaceful Happiness
Y Admits Sinners Happy Light
Z Agitates Charitables Excellent Exultation
& Develops Virtuous Uplifting Pleasures
9th 10th 11th 12th
A (Together with in Heavens Majesty Incomprehen- his Saints) sible
B Servants Ever and Ever Goodness God
C Loved Without end Kindliness Creator
D Saved In one Infinity Sapience Favour
E Beatified Perpetuity Charity Jesus
F Elected Sempiternity Power Transformator
G Confessors Enduring Infinity Dominator
H Apostles Incessantly Sublimity Preservator
IJ Evangelists Irreversible Benignity Immortal
K Martyrs Eternally Commiseration Supreme
L Angels In Glory Excellence Mighty
M Archangels In the Light Pity Omnipotent
N Dominions In Paradise Clemency Ineffable
O Proselytes Always Mercy Redemtor
P Disciples In divinity Divinity Sempiternal
Q Deified In Deity Deity Governor
R Ministers In felicity Omnipotence Rector
S Sanctified In his reign Virtue Sovereign
T Predestined in His Kingdom Love Invincible
UVW Preferred in beatitude Perfection Puissant
X Prophets in his vision Force Merciful
Y Patriarches in his magnif- Magnificence All Powerful cence
Z Cherubs to the Throne Grandeur Magnificent
& Professors in all Eternity Favour Sanctified
13th 14th
A Sincerely Preached
B Really Announced
C Saintly Published
D Evangelically Revealed
E Devotedly Denounced
F Intelligibly Acclaimed
G Evidently Exalted
H Publicly Sermoned
IJ Faithfully Interpreted
K Ardently Reported
L Constantly Narrated
M Sagely Served
26
N Carefully Praised
O Virtuously Recited
P Catholically Pronounced
Q Cordially Repeated
R Reverently Treated
S Theologically Speculated
T Justly Collated
UVW Divinely Spread
X Learnedly Cognitized
Y Entirely Recognized
Z Studiously Contemplated
& Spiritually Produced
Amen
Example:
Plain Tekst: 'DO NOT USE BEARER.'
Write this, spacing the letters and number them:
D O N O T
1 2 3 4 5
(The) King Triumphant Blessing (the) Bodies Manifests (to the )
U S E B
6 7 8 9
Catholics Pure Consolation (together with) His Servants
E A R E R
10 11 12 13 14
(in) Perpetuity The Majesty (of the) Rector Devotedly Treated.
AMEN
Look in the first alphabet for the word under 'D' - 'King' which you write under the first letter; look in the second alphabet for the word for 'O' - 'Triumphant', and in the third alphabet for the word 'N' - 'Blessing', and so on.
The encoded mesage will rea:
'The King Triumphant blessing the Bodies manifestts to the Catholics pure consolation tohether with his Servants in Perpetuity the Majesty of the Rector Devotedly treated. Amen.'
which sounds like a passage from a sermon or an old religious book.
The second part of the Book gave curious occult words
27
as code words, and the third was just a cabbalistic hotch-potch. But all these parts were interlarded with whole pages hinting at wonderful revelations to be made to those who persevered in studying the mysteries set forth in the book.
Unfortunately for himself, in the last part Tritheim alluded to a number of extraordinary characters which he called 'Spiritus nocturni' and 'diurni', and it was through this that he was accused of practising black magic.
The second part of Tritheim's book contains a great number of substitution alphabbets, using numerals ingeniously constructed in the form of disks (his printing costs must have been terrific). These were probably in use at the time and were merely collected. Tritheim seems to have been the first to arrange them systematically.
The unfortunate thing about his codes in the first book is that the coded message requires as many words as there are letters in the plain tekst, which makes the cryptogram exceedingly long. On the other hand, it is certainly hard to decipher, especially as the alphabets can easily be mixed up, in which case deciphering is almost impossible. The use of a new code word for each letter renders the frequency-table method useless.
The next mention of cryptography occurs in the famous book De Subtilitate by Jerome Cardan, who was born in Italy in 1501. Cardan was one of the most erudite and versatile natural philosophers of the Renaissance. By profession a physician, he was learned in all sciences, mathematics, theology, and astrology. His services as a doctor were at one time sought all over Europe by kings, dukes, and cardinals, and he lectured at the Universities of Bologna and Pavia. He achieved notoriety by his casting of the horoscope of Jesus Christ, to do which, in those days, required a good deal of courage as well as an established position.
28
He invented what is known as the trellis cipher which he himself describes as follows:
'Obtain twi identical pieces of parchment and cut holes in them of similar dimensions to the usual size of your letters, some of these to hold seven, three, eight or ten, enough to contain about one hunddred and twenty letters in all. One of the parchments you give to your correspondent and the other keep yourself. When you wish to communicate your secret you write what you want, taking care to put down one sentence only on each sheet though the holes on the parchment, and then you must try and compose an innocent-looking message to fill in the gaps. Be careful to avoid any suspicion, and preserve the continuity of your subject throughout the letter. When your friend receives it, all he has to do is to cover the sheet with his second parchment, and your secret message will immediately appear.'
This cipher, in several disguises, has been claimed as a new invention by various ingenious people ever since.
Italy in the sixteenth century was a hot-bed of political intrigue. The Council of Ten, the Borgias, the Visconti, and the Farnese were fighting, spying, poisoning, and scheming in a struggle for power. Obviously the had to communicate with each other secretly, and, as is often the case, the demand created the supply.
Ant Italian, Giovanni Baptista della Porta, of Naples, also a prolific writer on a variety of subjects, covering all the learning of his age, wrote a number of books in which he laid the foundations of a scientific approach to subjects which interest mankind even to-day.
His book on Human Physiognomy gave ideas to the Swiss writer Lavater, and in Human Magic there are important observations on mirrors and reflections of light rays, etc. He also wrote a volume on ciphers in which he mentions secret communications by means of bells, gestures, signs, torches, and invisible writing, and mentions a variation of the ancient skytale method which is rather ingenious.
29
Wind a thread roud a stick, the htread lying quite close to each other, and then write the secret message in ink across the thread, having first steeped the thread in alum water to prevent the ink from spreading. Then unwind the thread ansd send it either round a parcel or in a ball to your friend. No one will take notice of the markings on the thread, and only when it is wound round a stick with the same diameter as the Original one will the message be read.
Porta also gave a lot of advice on how to decipher secret messages, but nowadays his ideas are of no use. Nevertheless he was the first to point out the frequency of certain vowels, and the importance of differentiating between vowels and consonants, double letters, e.g., ll, tt, cc, etc.; and finally he proposed the use of an alphabet.
[Afb]
The nine groups were distinguished by nine fundamental characters, thus
[Afb']
These characters stood for the first letters in the groups; the second and third letters were indicated by the addition of dots in the angles of the characters. Thus ... - C and ... - N; ... - H, ... - S.
This artificial alphabet, with different modifications, was used extensively in Tudor times. Cardinal Wolsey used a modification of it in 1527 when he was English Ambassador in Vienna.
Porta devised yet another cipher, which is one of the most complicated of the early ones. It was probably the first cipher
30
invented with which a key word was used, and this table shows it:
| AB |
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 |
| CD |
a b c d e f g h i j k l m
z n o p q r s t u v w x y |
| EF |
a b c d e f g h i j k l m
y z n o p q r s t u v w x |
| GH |
a b c d e f g h i j k l m
x y z n o p q r s t u v w |
| IJ |
a b c d e f g h i j k l m
w x y z n o p q r s t u v |
| KL |
a b c d e f g h i j k l m
v w x y z n o p q r s t u |
| MN |
a b c d e f g h i j k l m
u v w x y z n o p q r s t |
| OP |
a b c d e f g h i j k l m
t u v w x y z n o p q r s |
| QR |
a b c d e f g h i j k l m
s t u v w x y z n o p q r |
| ST |
a b c d e f g h i j k l m
r s t u v w x y z n o p q |
| UV |
a b c d e f g h i j k l m
q r s t u v w x y z n o p |
| WX |
a b c d e f g h i j k l m
p q r s t u v w x y z n o |
| YZ |
a b c d e f g h i j k l m
o p q r s t u v w x y z n |
31
Each pair of capitals in the left hand column together control the alphabet ranged in two lines to their right. The keyword is formed from these capitals and its letters in succession indicate the alphabets selected. For instance, to cipher the word 'red' by means of the code word 'car', you find the capital letter 'C' in the left-hand column, and look along the line to the right until you find the small letter 'r''. The ciphered letter is the one that appears above or below this letter - in this case 'f'. Repeat the process with the other two letters and the result is 'frv'.
It stands to reason that to decipher, the same method is adopted, if NOT is the keyword 'vtu'' - bad.
Porta suffered almost immediately from a copyist. A French diplomat, Blaise de Vigénère, author of several books on occult subjects as well as an alchemist and astrologer (the usual combination in the sixteenth century), wrote an Occult Treatise on Letters, in which, between futile cabbalistic dreams, he mentions a good many of Porta's ideas, though they were not as precisely and clearly expressed as in the Italian book. Nevertheless Vigénère produced an ingenious code, which is very similar to one suggested by Bacon and still used to-day. This is as follows:
|
AA |
BB |
CC |
AB |
AC |
BC |
CB |
| A |
a |
d |
g |
k |
o |
v |
u |
| B |
l |
e |
h |
m |
p |
s |
x |
| C |
c |
f |
i |
n |
g |
t |
z |
The small letters are the plain tekst, the capitals the cipher.
32
To cipher 'the foe' , one finds the small letter 't', and writes down first the two capitals in the top horizontal line and then the single capital in the left-hand column; thus 't' - BCC, and 'the foe' - BCC CCB BBB BBC ACA BBB.
For greater secrecy the coded message should be sent in groups of five letters, and should not show the actual length of words as this would help to decipher the message.
Vigénère also arranged a modified version of Porta's multiple alphabet cipher which is given on p. 33. Below is an example showing the use of a code word TROY. The procedure is similar to that used with Porta's table, only each letter of the key word (on the side of the table) gives the entire cipher alphabet, the plain tekst alphabet being on the first top line (printed in capitals).
Key word: TROY
Message (plain text): 'FOOD SUPPLIES RUNNING OUT''
Keyword: TROY TROYTROY TROYTRO YTR
Message written for enciphering: FOOD SUPPLIES RUNNING OUT
Cipher: AGDB NNENFBSQ MNCLDFU MPM
Cryptogram: AGDBN NENFB SQMNC LDFUM PMZYX
Almost at the same time a noble duke - the Duke of Brunswick-Luneburg - entered the field of secret writing, and he published a book called Cryptographiae under the pseudonym of Gustavus Selenus. This was in itself a cryptogram: SELENE means 'moon' in Greek and Luneburg means 'moontown'', while Gustavus was an anagram of his Christian name, Augustus.
33
[Afb.]
Vigénère Code
He was devoted to the study of the occult sciences and cabbalism and wrote a number of books on various subjects, amongst others, Systema integrum Cryptographiae, in which
[N]
34
he dealt with codes and ciphers, borrowing mostly from the Abbot Tritheim; but he was the first to make mention of a vowel cipher, which is not without its merits.
A-b A-b A-p
E-c E-k E-q
I-d I-l I-r
O-f O-m O-s
U-g U-n U-t
The idea is that if there are two or three 'e''s for instance in a word, the C, K, and Q shall be used after another, and so for all vowels, while the consonants are first replaced by corresponding vowels in the cipher, thus:
EMPEROR OF AUSTRIA would be:
c o a k i f i f o b g o u i d h
This could be further improved by placing dummy letters between words and writing them either in groups of five or continuously, the idea being to prevent deciphering by means of a frequency yable, which starts with vowels. The duke also mentioned an invented alphabet attributed traditionally to Solomon. Here it is:
[Afb.]
35
The name of Francis Bacon has in recent times been associated with the subject of codes and ciphers by reason of the Shakespeare-Bacon controversy; but, apart from this, Bacon - philosopher, scientist, essayist, poet, and Lord Chancellor of England - made his own contribution to the science of codes and ciphers. One of his codes has been already referred to; here is another:
[Afb.]
When the Shakespeare-Bacon controversy was at its heighth at the end of the nineteenth century, a Mrs. Gallup, reading this cipher, wondered why Bacon should invent it and use it in the printing of books. Ciphers generally are made to be used at a moment's notice, and he himself says that ciphers should be easy to read and write; but to set up a book in several different types is an expensive process, and nobody would do it without reason. This reason Mrs Gallup set to work to find. She carefully studied the first edition of one of Bacon's early works, and on the title-page, hidden under two sets of italics, she discovered the name of William Rowley - Bacon's chief secretary.
She went on with the work of deciphering and wrote several books on the subject. She claimed to have found this cipher not only in the books of Bacon himself but also in the early editions of works by Greene, Marlowe, Shakespeare,
36
and Ben Jonson. But her startling claim was that the enciphered story thus hidden proved that Bacon was the author of all Shakespeare' plays, and, furthermore, that he was the eldest son of Queen Elizabeth by her first marriage to the Earl of Leicester, which took place while she was imprisoned in the Tower by Queen Mary.
In any case, in the first edition of Shakespeare's works, published by Isaac Jaggard in 1623, Mrs. Gallup claims to have found the following message, enciphered in the 'L. Digges' Poem.
'Francis of Verulam is the author of all the plays heretofore published by Marlowe, Greene, Peele and Shakespeare, and of the twenty-two (plays) now put out for the first time. Some are altered to continue his history. Fr. St. A.'
The difference between the two sets of italic founts being very small, it is an extremely difficult matter to decipher these messages. Butfor us the important point is that cipher was employed as a weapon in this controversy.
Francis Bacon himself, however, introduced into his charge of treason against the Earl of Somerset the fact that the accused communicated with his friends in cipher, 'a process reserved for Kings and Princes in their affairs of State'.
Another famous trial, where the charges brought against the accused were based on deciphered correspondence, was beheaded because of the letters he wrote to his wife in cipher, denouncing his enemies and plotting against the Parliament. These papers were captured at the battle of Naseby andwere deciphered by a Dr. Wallis.
The beginning of a more scientific approach to the business of codes and ciphers becomes apparent about this time.
37
It is exemplified in the case of the Spaniards, who wished to maintain close relations between the different parts of their vast and scattered empire. At that time the Spanish dominions included the Netherlands, part of Italy, and all the Nrew Wrorld in South America. Naturally enough they desired to keep their internal communications private, and invented for that purpose a special alphabet composed of fifty arbitrary signs. Though this was of the greatest use to them during the League Wars an their constant enmity with France, it was bevertheless not quite as efficient as they had imganined. Henry IV intercepted a few of their dispatches, and asked a well-known professor of geometry - Viette - to see if he could decipher them. The mathematician , acquainted with the 'frequency law', succeeded in finding the key to the secret alphabet in all its variations, and for two years the French Foreign Office deciphered Spanish dispatches, which they seized and afterwards sent on so that the Spaniards would suspect nothing. Eventually, of course, the Spanish government found out what was going on and accused Henry IV, who was a Protestant, of having magicians at his service and of invoking the help of the devil in order to find our Spanish cryptographic secrets, maintaining that he could only have deciphered the message by calling up the spirits of those who had known the cipher during their earthly life.
But times were changed. Rome was acquainted whith the works of Cardan and Porta and the Pope had a sense of humour. He forwarded the complaint to be investigated by a commission of cardinals 'with urgent recommendation'. The cardinals were not slow to understand the Pope's intention, and their investigations into this grave matter are not yet completed.
At the beginning of the eighteenth century the Elector of
38
Brandenburg - Frederick III - decided that he ought to be crowned king and convert his duchy into the kingdom of Prussia. But without the consent and support of the Emperor - the head of the Holy Roman Empire and titular ruler of all Germany - it was almost impossible to accomplish this. Negotiations were accordingly started with the Court of Vienna, the capital of the Empire, which was not too keen to grant such an exceptional honour to a growing duchy lest it become too powerful, and for years the negotiations languished without any result. The Minister of the duchy at the Viennese Court, Baron Bartholdi, used for his correspondence with Frederick III a figure cipher, in which well-known persons constantly mentioned in dispatches were specially numbered - a common practice.
Amongst these was a Jesuit, Father Wolf, the confessor of the Emperor's Austrian Ambassador in Berlin; he was numbered 116. He played a significant role in the petty diplomatic intrigues of the times, and was a personage of some importance. Frederick III was number 24 and the Emperor number 110.
One day Bartholdi sent a message from Vienna saying that 24 (thee Elector Frederick) should write a personal letter in his own handwriting to 110 (the Emperor); but the letter was hastily written, and the '0' of 110 was taken for 6, reading - Father Wolf.
Frederick did not hesitate, and wrote in his own hand a letter to Father Wolf, stating his reasons for wishing to convert the duchy into a kingdom, and requesting the Jesuit to help him accomplish his project.
Father Wolf was as surprised as he was flattered to receive such an important communication. He decided to do all he could through the powerful order of Jesuits to help a prince who requested the help of the church in the furtherance of
39
his plans. He wrote a letter to the Father Confessor of the Emperor, and he communicated with the General of the Society of Jesus in Rome; the church decided in favour of the Elector; pressure was brought to bear on the Emperor, and, thanks to the Original misreading of the cipher, Frederick quickly obtained the consent of the Austrian Court and was crowned King of Prussia.
History does not relate what form the gratitude of Frederick took to the poor Jesuit Father who really got him his crown, but one hopes that he was suitably rewarded.
History has many other celebrated names of men who used cipher, among the Cardinal Richelieu in France.
The Duc de Broglie mentions in his book, The King's Secret, a jargon cipher - a kind of dictionary code - which Richelieu used in his private correspondence. A similar one was employed also by Mazarin. In this jargon 'Coxcomb' stood for Grand Duke, 'Garden' for Rome, 'Rose' for the Pope, 'Greyhound' for the Emperor, "May Lily' for Cardinal de Medici, 'Conice' for the Queen. At that time the substitution of the names of flowers for the names of real people was quite common. Richelieu himself gave such a code in his own journal.
But for his state correspondence with Louis XIV he occasionally used a very clever form of transposition cipher with a multiple key. Here is an example:
Key: 2741635 ; 15243 ; 671852494 ; 0728615943
Plain text: LETTER SENT TO THE
Key: 274163 5152 43 671
Plain text: EMPEROR GIVING FULL
Key: 8524930 728615 9432
Plain tekst: DETAILS
Key: 7416351
40
1st keyword 2nd keyword 3rd keyword
1234567 12345 123456789
Cipher TLRTSEE ETOTN EPOEMTHER
4th keyword 1st keyword again
1234567890 1234567
Cipher: NILUGIGVEFR TLIESAD, etc.
Cryptogram: TLRTSEE ETOTN EPOEMTHER NILUGIGVFR TLIESAS, etc.
|
|
III. SIGNALS, SIGNS AND SECRET LANGUAGES
48
Before we go to the more complicated codes of to-day, let us examine a method of signalling which, in different forms, has been in use since the first hollowed tree-trunk was launched on the sea.
The first actual records of signals at sea are mentioned in Greek mythology when Aegeus sent his son to Crete and agreed with him beforehand that a white flag should be displayed if he reached safely. Virgil also mentions that Agamemnon signalled from his ship to Sinon in the citadel - but these were probably not signals as we understand them. Certain flags, torches, or a sound may have conveyed an agreed meaning, but not letters or words. We have already mentioned Polyvbius' system of torch signals in the historical survey in the first chapter; this was the first signalling code ever invented, but we must note also in this connexion a code of Joachimus Fortius whereby three torches were used in this manner:
One torch alone - first 8 letters of the alphabet - ABCDEFGH
Two torches together - next 8 letters of the alphabet - IKLMNOPQ
Three torches together - the rest of the alphabet - RSTUVWXXYZ.
and to differentiate these three groups - one light used once stood for the letter 'A', one light twice for 'B', two lights for 'T', two lights twice for 'K'', and so on.
49
Before the advent of wireless, ships at sea relied very much more then they do now on their visual signals. This is the numerical code of light signals which is still used very frequently, with four lights only, placed on the foremast.
[Afb.]
During the days of sailing-ships signalling lights were used placed on te cross made by the mast and spars, i.e., vertically and horizontally to each other. But now this has been superseded by the Morse code, the flashes of which are made by powerful electrical projectors.
50
[Afb.]
Semaphore, however, is still in use, and most people are familiar with this form of signalling, which is done either with flags or signal arms. The arms move in a circle which is divided into eigth parts, each of 45 degrees; seven of the parts are the letters A - G, and the eigth position is where the arms are placed when not in use.
Procedure Signals and Special Signs
Heraldry
The fact that heraldry should come into a book dealing with codes and ciphers may seem a little strange, but is not heraldry an abbreviated and, to most people, a secret way of giving a person's family antecedents ? It is a more beautiful and picturesque form of the laundry mark.
It may also sound rather strange to say that the use of flags
52
had its origin in the belief in the transmigration of souls; nevertheless, this appears to be true. Certain tribes of semi-savage people believed that when they died their souls would live again in the form of an animal (primitive totemism), and to ebe sure that they returned into that of a worthy animal, they often wore amulets, made out of rough pottery or clay, about their person during their lifetime. The stories of the werewolf remind us of this old belief, and the practice of wearing lucky charms and carrying mascotrs on motorcars may be regarded as a survival of it.
IV. COMMERCIAL CODES
70
As has probably been noticed, codes and ciphers are no longer used only by kings and generals, or by clandestine lovers with irate fathers; with growing commercialism there have sprung up codes to help trade.
V. MILITARY CODES AND CIPHERS
89
Cryptography is used to a large extent by the army, which not only provides some training for cipher officers but has evolved certain methods suitable for military purposes.
70
Lloyd’s
In 1688 Edward Lloyd ran a coffee-house in Tower Street, London. An enterprising man, he found that several brokers used to discuss their business over coffee. To sell more coffee, he decided he must make things easier for them. He instituted a blackboard, and then a weekly bulletin of shipping information. More independent brokers came and consumed his coffee while doing their business. He later moved his coffee house to Lombard street, in the very center of the old city of London frequented by merchants of the highest class. It was not until 1774, with the rapid increase of marine insurance business, a committee was set up and a constitution formed which has remained practically unaltered to the present day. There is no longer a Lloyds' coffee-house, yet the name is preserved, and Lloyds'is known all over the world as the center of the Marine Insurance business.
Lloyds devised a method of signalling between sea and shore, so that advance news of ships and cargoes might be received. A primitive projector was set up and a system of light signals based on the Polybius' system was started. It was this that gave rise later on to the use of codes for commercial purposes; and apart from the Venetian merchants in the eighteenth century, Lloydssignals were the first to come into general use. In 1794 in Europe, a system of rapid communications known as ' aerial telegraphy', employing semaphores on high towers visible at considerable distances, was instituted. Whole phrases or sentences could be expressed by one group of signals.
In 1825 codes employing figure groups were in common use. In 1845 the Telegraphic Vocabulary Code was used between Liverpool and Holyhead for the semaphore telegraph. In this code there appear words, phrases, long sentences, each represented by groups of one to four digits.
In England the earliest practical trial of electric telegraphy was made in 1837 on the London and North Western Railway, and the first public line, under Wheatstone and Coke Patents, was laid from Paddington to Slough on the Great Western railway in 1843. [DAGA]
In America, in 1860, Brewell published his Mercantile Cipher for condensing telegrams, in which English dictionary words were employed, and in which we find a fairly complete vocabulary, arranged under captions.
The Concise Oxford Dictionary, 3rd edition, 1934.
VI. TYPES OF CODES AND CIPHERS
103
In order to become efficient in the knowledge of codes and ciphers some system of classification must be adopted. As we have already seen, many methods have been employed at different times to preserve secrets from inquiring eyes, and cryptographers throughout the ages have displayed remarkable ingenuity; nevertheless, we can make a few general divisions of ciphers.
114
4. Dictionary Code Systems
These are highly specialized forms of substitution systems, and involve the use of modified dictionaries known as code books. The commercial uses of these codes were explained in Chapter IV, but the code books used by the Foreign Office or for military purposes represent a greater condensation - a single code group may represent a long phrase. The average condensation of a diplomatic code is often 1:5 while in a commercial code it is only 1:3.
These groups, as we have seen, are all pronounceable artificial words, such as ‘ABACA’ for instance in a commercial code or ‘EXA’ in a Diplomatic code, and accasionally syllables such as ‘BA’, as we have seen in examples of Marconi codes.
As it is difficult to safeguard against the loss of code books, which have to be printed in great numbers, they do not afford such security as ciphers, although, of course, so long as they are kept secret, they are very good.
A story about a dictionary code is told by Mr. J.C.H. Macbeth. During the Russo-Turkish war in 1877, the Ottoman Field-Marshal Osman Pasha entrusted one of his generals, Selim Pasha, with a confidential mission. It so happened that Selim was the officer responsible for ciphering, and being prudent, he always kept the code on his person. Unfortunately, he departed so promptly on his
115
mission that he forgot to leave the volume with his chief. And the latter, during the whole time of his Adjutant’s absence, saw a pile of ciphered telegrams from Constantinople accumulate on his table without being able to read or reply to them.
Codes, of course, can be used in conjunction with ciphers, and an enciphered code of this kind should be very difficult to break; but the work and time involved in making such a combination would be a serious objection. Speed of encoding and decoding is essential.
One of the ways in which ordinary dictionaries can be used is first to agree on a certain edition, say, for instance, the Concise Oxford Dictionary, current edition, by Fowler and Le Mesurier, and then to give only the number of the page, and the number of the word down the page. ‘Reunion Berlin to-morrow’ would be enciphered thus:
1006 (page no,) 12 (word no.) - Reunion.
0104 (pages with fewer than 4 numbers would have ‘0' added in front to keep to the uniformity) 17 (word no.) - Berlin
1291-08 (on the same principle) - To-morrow
And the cryptogram would read: ‘100612 010417 129108'
These figures, if greater secrecy is required, could again be enciphered and thus converted into letters by means of an agreed cipher. For that purpose it would be better to arrange for the second operation. Divide the figures into pairs and then convert them into letters by means of the table given on p. 116.
116
|
|
2 |
n |
d |
|
F |
i |
g |
u |
r |
e |
|
|
1 |
|
1 |
3 |
5 |
2 |
4 |
9 |
7 |
8 |
6 |
0 |
|
s |
9 |
AN |
DA |
HN |
JT |
MB |
KC |
GF |
ES |
BZ |
ZA |
|
t |
2 |
CK |
AO |
DB |
HO |
JS |
GE |
ER |
BY |
FR |
YB |
|
|
7 |
IR |
CJ |
AP |
DC |
GD |
BQ |
BT |
FQ |
LH |
VA |
|
f |
4 |
MC |
IY |
CI |
AR |
DD |
BS |
FP |
LI |
NL |
VB |
|
i |
8 |
MA |
KB |
GC |
CG |
AS |
DF |
HP |
JU |
OB |
VC |
|
g |
1 |
KA |
GB |
EP |
BR |
CE |
AT |
DG |
HQ |
JQ |
TZ |
|
u |
5 |
GA |
EO |
BP |
FO |
IX |
CC |
AX |
DH |
HR |
TY |
|
r |
3 |
EN |
BO |
FN |
LJ |
NK |
IZ |
CB |
AY |
DJ |
SB |
|
e |
6 |
BN |
FM |
LK |
NJ |
OA |
OC |
IV |
CB ? |
AZ |
QA |
|
|
0 |
XY |
YA |
BY |
YB |
XC |
XE |
YD |
YE |
YX |
QC |
NULLS: WA, WE, W, to end message in groups of five letters.
Note
I have analyzed the pattern of this table as follows:
116
|
|
|
A |
|
|
|
|
|
|
|
|
|
|
|
|
1 |
3 |
5 |
2 |
4 |
9 |
7 |
8 |
6 |
0 |
|
|
9 |
AN |
DA |
HN |
JT |
MB |
KC |
GF |
ES |
BZ |
ZA |
|
|
2 |
CK |
AO |
DB |
HO |
JS |
GE |
ER |
BY |
FR |
YB ? |
|
|
7 |
IR |
CJ |
AP |
DC |
GD |
BQ |
BT |
FQ |
LH |
VA |
|
|
4 |
MC |
IY |
CI |
AR |
DD |
BS |
FP |
LI |
NL |
VB |
|
|
8 |
MA |
KB |
GC |
CG |
AS |
DF |
HP |
JU |
OB |
VC |
|
|
1 |
KA |
GB |
EP |
BR |
CE |
AT |
DG |
HQ |
JQ |
TZ |
|
|
5 |
GA |
EO |
BP |
FO |
IX |
CC |
AX |
DH |
HR |
TY |
|
|
3 |
EN |
BO |
FN |
LJ |
NK |
IZ |
CB |
AY |
DJ |
SB |
|
|
6 |
BN |
FM |
LK |
NJ |
OA |
OC |
IV |
CB ? |
AZ |
QA |
|
|
0 |
XY |
YA |
BY ? |
YB ? |
XC |
XE |
YD |
YE |
YX |
QC | NULLS: WA, WE, W, to end message in groups of five letters.
1st diagonal top-down: AN-AO-AP-AR-AS-AT-[AU]-[AV]-AX-AY-AZ
2nd diagonal down-top: BN-BO-BP-[BQ]-BR-BS-BT-[BU]-[BV]-[BX]-BY-BZ
3rd diagonal down-top: CA?-CB-CC-[CD]-CE-[CF]-CG-[CH]-CI-CJ-CK
4th diagonal top-down: DA-DB-DC-DD-[DE]-DF-DG-DH-[DI]-DJ
5th diagonal down-top: EN-EO-EP-[EQ]-ER-ES
6th diagonal down-top: FM-FN-FO-FP-FQ-FR
7th diagonal down-top: GA-GB-GC-GD-GE-GF
8th diagonal top down: HN-HO-HP-HQ-HR
9th diagonal mixed: IR-IY-IX-IZ-IV
10th diagonal mixed: JT-JS-JU-JQ
11th diagonal down-top: KA-KB-KC
12th diagonal top-down: LH-LI-LJ-LK
13th diagonal down-up: MA-MB
0 lines: horizontal X, Y, vertical V, Y, Z combinations.
Mistakes ? : BY (2x), CB 2x), YB (2x), CB (2x). CB in 68 must be a serious mistake for CA, because it falls in the diagonal context.
The numbers enciphered into letters would be: TZYXBR XYXCDG BRANYE
and the cryptogram for transmission would be TZYXB RXYXC DGBRA NYEWA
The suggested cipher can easily be arranged to make pronounceable words, suitable for telegraph transmission.
Certain dictionaries have been issued which give the two columns on each page with words directly opposite to each other. Then it is possible to give the word opposite the one we realy mean, or a word which is 5 or 3 or 10 places either above or below the one we want to encode. Codes of this kind can be solved without difficulty, and I will give an example in the next chapter.
VII. METHODS OF DECIPHERING
121
There is a very sad story which shows that the lack of knowledge of the elementary deciphering principles may sometimes play an important role in one's life.
129
[A FRIEND'S CRYPTOGRAM]
I asked a friend to give me a cryptogram to decipher without letting me know anything about the kind of cipher that was used. As I think it was quite instructive from the point of view of the mistakes I made, and the mistakes made also by my friend in taking the cipher down, I shall give it here. This is what I received:
CDDBC ECBCE BBEBD ABCCB BDBAB CCDCD BCDDE CABCB DDDAA CABCE AAABD EBCED CBCCD AEBDC BAAEA BECDD BDCCB CEEAB DADEA DCAAD EACAB DCBDC BAABD CACED CBABC DDCDB DDCBE BCDCB EBCAA BDACC DDBBB CEAAC DBDCD DBCED CAECA CEDC
It will be noticed at once that only the first five letters of the alphabet repeat themselves: 'abcde'. By certain repetitions, etc., in the crypt I felt, rather than was able to prove, that it was a two-letter cipher.
This is the principle of such a cipher:
2nd letter
A B C D E
---------------------
A a b c d e
B f g h ij k
C l m n o p
D q r s t u
E v w x y z
Exemple: ENEMY would e: AE for 'E', CC for 'N', exc., making complete crypt: AE, GC, AE, CB, ED.
The letters are seldom put in such a straightforward alphabetical order; both the plain text and the cipher letters (capitals) would be all jumbled together and not put in any obvious order. In any case, the first procedure was to
130
separate the crypt in two-letter groups with red pencil, and then make up a frequency table, thus:
|
AA occurs 6 times
AB occurs twice
AC occurs 3 times
AD occurs once
AE occurs 3 times |
BA occurs twice
BB occurs twice
BC occurs 7 times
BD occurs 9 times
BE occurs 3 times |
CA occurs twice
CB occurs 5 times
CC occurs 0 times
CD occurs 9 times
CE occurs 10 times |
|
DA occurs twice
DB occurs 4 times
DC occurs 9 times
DD occurs 3 times
DE occurs once |
EA occurs 3 times
EB occurs twice
ECoccurs once
ED occurs 0 times
EE occurs 0 times
|
|
According to the table of frequencies, the group CE must be equivalent to the plain text 'E'. So we write it above the crypt in red pencil in appropriate places, this way:
E E E E
CDDBC ECBCE BBEBD ABCCB BDBAB CCDCD BCDDE CABCB DDDAA CABCE AAABD EBCED CBCCD
E E
AEBDC BAAEA BECDDBDCCB CEEAB DADEA DCAAD EACAB DCBDC BAABD CACED CBABC DDCDB
E E E
DDCBE BCDCB EBCAA BDACC DDB BBCEAAC DBDCD DBCED CAECA CEDC
On examing the crypt now we observe the repetition of the 'CD, DB' four times. This bigram forms on two occasions part of the trigram CD, DB, CE, and the last group CE we know to be 'E'.
131
occasions part of the trigram CD, DB, CE, and the last groep CE we know to be 'E'. The commonest bigram is ''th' and the most used trigram is 'the', so it is quite easy to deduce that 'CD, DB' stand for 'th'. Now take a red pencil and mark 'th' over the crypt yourself.
It was here that I went wrong. I looked at the line before last of this group
AA BD AC CD DB BB CE AA CD
Fitting in the letters we know, it looks like this:
AA, BD, AC, th, BB, e AA t,
and I jumped to the conclusion that it was 'North West'. The group BD occurs nine times, and the table of frequencies gives 'O' as being very near to 'E' in recurrence; but 'AA' stood in the beginning for 'N' in 'north' and for 'S' in 'west', and that did not fit. But 'South West' would fit very well in all cases.
So I started to mark the appropriate letters in the cipher with the plain tekst letters thus derived, when I observed the following group on the fourth line:
CD DB DC CE CE
or if I put in the values which have been already deduced - 't, h, DC, e,e'. There is only one word of five letters 'TH?EE' and this is 'three'. So we have one more letter to add to the plain tekst - 'R'.
So let us see once more how the crypt reads now, when we fill in the plain texts values we know.
AA - S CE-E AC-U DC-R
CD - T BB-W BD -O DB-H
The - CB, e, w, EB, DA, BC, CB, o, BA, BC, t, t, BC, DD, BC, AE, CB, DD, DA, u, AB, e, s, o, EB, e, r, BC, t, AE, o, CB, s, EA, BE, - three - EA, o, AD, EA, r, s, DE, u, AB, r,, o, CB, s, o, CA, e, r, BA, BC, DD, t, o, r, BE, r, BE, BC, south - west - o, the, r, AE, CA, e, r.
132
My attention was drawn next to the following groups: 'o,CA,e, r,' and the last group: 'r, AE, CA, e, r.' If the first be taken as 'over' then the value of CA - V, in second group we have 'rAEver' which was 'river' and here I come to the first mistake made by the 'encipherer'. The 'F'' was omitted from the word 'of'.
This made me realize that probably other omissions would be met with in the crypt; but in any case we had two more letters to add to our plain tekst values: CA-V and AE-I.
Now once again take a red pencil and mark over the cryptogram AE-I, as there are no more 'CA' groups to mark. At once this group 'o, EB, e, r, BC, t, i, o, CB, s' jumps to the eye. If we write it this way 'o?er?tio?s' - it can only be 'operations'. This will give us the values of EB - p, BC-a, CB-n.
If we write these new values in, the crypt will now look as follows:
The new p (DA) an o(BA) arta(DD) (EC) in (DD) (DA) u (AB)es operations (EA) (BE) three (EA)o (AD) (EA) rs (DE) u (AB)rons over (BA) a(DD)tor (BE) ar(BE) a south west or the river.
The group 'atta(DD) (EC)' can only mean 'attack' or 'attacking', but the value of 'DD' being 'C' in this case, it precludes 'attacking', and the following word 'inc(DA)u (AB)es' looks like 'includes'.
The new plan o(BA) attack includes operations (EA) (BE) three (EA)o(AD) (EA) r s (DE) (drons over (BA)actor(BE)ar(BE)a south west on the river.
'plan o(BA) attack'' gives the only possible value to 'BA-F, and 'operations (EA) (BE) three ...' must be read as 'EA, BE - by'.
Now if you bear in mind that we have already found one '
133
mistake in ' south west o? the river' - 'F'' in the 'OF'' missing - you must be prepared for other mistakes. There are two missing letters and one misspelling; nevertheless, if the last three values found are written in, the sense of the message is apparent.
It is very common in deciphering to come across mistakes either by the encipherer or by telegraphists, signallers, or others concerned with the transmission of the cipher. If the above example had been without errors the deciphering would have been easier, but even as it was we succeeded in getting the sense completely.
140
Dictionary Code
I have already described the dictionary code, and it is clear that it would be the most difficult one to decipher. Yet, Mr. Mansfield, an Australian criminologist of repute, has suggested some extremely interesting principles for the solving of such codes, and has even calculated dictionary progressive lists, giving numbers of words beginning with any two letters in dictionaries of 10, 20, 30, 40, 50, 60, 80 and 100 thousand words.
Mansfield’s methods are very simple and ingenious. They are better demonstrated by an actual example. Let us take the case, where all the words in a certain dictionary are numbered, and these numbers are transmitted as a code.
Here is a message
55381 42872 35284 [55]381* 45174 56037 55381 46882
23171 44234 55366 55381 00723 12050 61571 36173
55381 56442
* was 44381
Let us analyse this list, starting from the lowest numbers and finishing with the highest.
00723
12050
23171
35284
36173
42872
42872
44234
44234
45174
45174
46882
46882
55366
55366
55381 (5 times)
56037
56442
61571
Words beginning with XYZ are seldom used, so we can take it that the highest number indicates a word beginning wit a ‘W’ or a ‘T’. But the list of frequencies gives us the commonest intial group as ‘th’ or ‘the’, and if we can find any repetition of such nature, that will fix ‘T’ in the dictionary. The group 55381 occurs five times in the crypt, so most probably it is ‘the’.
The highest number after that is 61571, so that it should indicate a wordt beginning with a ‘W’. This gives us the clue to the probable number of words in the dictionary used for that code. It cannot be over 65,000 as XYZ words are very few, seldom over 3000. Even in the largest dictionaries 60,000-65,000 words are quite common.
According to Mr. Mansfield’s Progressive Dictionary we can now fix the probable first two letters of every word in the code. For instance, the second group 12050 will be between 11646 (terminating words beginning with ‘DA’) and 12850 (terminating those beginning with DE, so that it is certain to be a word beginning with DE. Using Mansfield’s table we obtain:
‘The re- of the ro- ro- the se- - ha- re- th- the re- de- - wa- ov- the to-
Now we can find in the dictionary the word ‘the’ (55381). We count back about twenty words for 55366 (th). This gives us an area covering the words ‘thane’, ‘thank’, ‘that’, ‘tharch’. We will try ‘that’ as the most likely out of these. We now consider the groups 56037 and 5644. Words beginning with ‘to’ start at 56037 and stop at 56466, so that we should be justified in assuming the first to be ‘to’, while for the second (56442) we count back from the end of the ‘to’ section and find that ‘town’ seems the most likely word.
Let us now consider the ‘r’ group, 42872, 44234, and 45174. The Re’s’ begin at 42574 according to Mansfield’s table, and counting 300 words from here gives us the following group to select from: ‘recline’, ‘recommend’, ‘recompose’, ‘reconnaissance’, ‘recoup’, and ‘recover’, and the word we choose is ‘reconnaissance.
The same proces brings us ‘revealed’ and ‘route’ for the other two.
What does the cipher look like with these few words put in ?
The reconnaissance of- the route to the se- ha- revealed that the ae- de- wa- ov- the town.
We apply the same process to ‘ae-‘ 00723, and we get straight on to ‘aeroplane’, while ‘de-‘ 12050 occuring one-quarter of the way from the end of DA to the end of DE brings us to DEF limited by ‘deface’ and ‘defy’, where only ‘defeat’, ‘defence’, ‘defend’, and ‘defensive’ are probable. We can select ‘aeroplane defensive’ as near our mark.
The same process brings us ‘sea’ for ‘se-‘ (46882) and ‘over’ for ‘ov-‘ 36173. The complete message reads now:
The reconnaisance ‘of-’ the route to the sea ha- revealed that the aeroplane(s) defensive wa- over the town.
We can easily see that ‘of-‘ is merely ‘of’, ‘ha-‘ is ‘has’, and ‘wa-‘ is ‘was’. The actual message differed only in two words: instead of ‘aeroplanes’ ‘defensive’ it was ‘aerial defence’, but the meaning was the same.
This application of the law of probability to dictionary codes is very interesting. The search in the area of possible words will at any rate give us the root of the plain text used in the actual code, and from that root we can usually deduce the whole of the meaning hidden by the code. (See Louis C.S. Mansfield: The Solution of Codes and Ciphers.)
LIST
The following lists may be of some use for deciphering purposes.
English diphtongs: ae, ai, ao, au, aw, sy, ay, ea, ee, ei, eo, eu, ew, ey, ia, ie, oa, oe, oo, ou, ow, ue, ui.
Triphthongs: eau, eou, ieu, iew, iou.
Double consonants: ss, tt, ll, mm, ff, bb, cc, dd, gg, nn, pp, rr, zz.
143
Letters used most: E, T, I, O, N, A, S, H. But only the frequency of E can be relied upon.
Letters used last: Z, Q, J, X, K, Y.
Words terminating in: T, N, S, Y, D, E, especially E.
E and O are the easiest to discover, E because it is most often used, and O because it is the oftenest used as part of a bigram; also because both E and O are about the only ones used doubled in the middle of words of four letters.
Monograms: A, I, and sometimes O.
Bigrams: ab, ah, ad, at, an; be, he, me, we, ge; if, in, is, it; do, go, lo, no, of, on, or, so, to, wo; up, us, ut; by, my, fy, ye.
Of these the order of frequency is: of and to, it, is, in, we he, as, or be, by, so and on.
If the first letter of a bigram is O, the second leter is usually F, N, or R.
If the first is T or S, the second is O.
If it is I, the second letter must be T, S, N, or F.
If it is W, H, or Y, the second is always E.
If the first is B, the second is E or Y.
If it is A, the second must be S, B, H, or T; most often S.
If two bigrams follow each other they will be TO DO, or DO NO.
If the first letter of two bigrams are the same, they will be IF IT, IN IT, IT In, IT IS, OR OF, OR ON, but the first letters will always be either I or O.
If the first letter of the first bigram is equal to the second letter of the second bigram, it will be OF SO, OF TO, TO IT.
If the second letter of the first bigram equals the first letter of the second it will be AS SO, IS SO.
144
If we find a word which has this ending: ... 3166 x 166, we can at once decide that the second double letters are SS, the first two LE, and that the two preceding the second lot of double letters are NE, such as in 'blamelessness', 'dreadfulness', 'needlesness'.
Trigrams: THE, AND, THA, [ING], HET, EDT, ENT, FOR, ION, TIO, NDE, HAS, MEN, NCE, OFT, STH.
Three-letter words: THE, END, FOR, ARE, NOT, BUT, HAD, HAS, YOU, WAS, HIS.
Four-letter words: THAT, WITH, HAVE, FROM, THEY, THEM, THIS, WHEN, WILL, OVER, BOTH.1.
1. Wikipedia
Most common bigrams (in order)
th, he, in, en, nt, re, er, an, ti, es, on, at, se, nd, or, ar, al, te, co, de, to, ra, et, ed, it, sa, em, ro.
Most common trigrams (in order)
the, and, tha, ent, ing, ion, tio, for, nde, has, nce, edt, tis, oft, sth, men
Solution of crypt, p. 128 (mono-alphabetic substitution):
General ordered the second brigade to attack at 3.30 a.m. the third brigade one hour later on the left, the fourth keep in reserve.
Solution of Vigénère cipher, p. 135 (keyword CID):
No one here has deciphered the three latest dispatches, please discontinue these ciphers as the ones used hitherto were better.
Here is a cryptogram upon which the reader is invited to test his skill.
75628 28591 62916 48164 91748 58464 74748 28483 81638 18174
74826 26475 83828 49175 74658 37575 75936 36565 81638 17585
75756 46282 92857 46382 75748 38165 81848 56485 64858 56382
72628 36281 81728 16463 75828 16483 63828 58163 63630 47481
91918 46385 84656 48565 62946 26285 91859 17491 72756 46575
71658 36264 74818 28462 82649 18193 65626 48484 91838 57491
81657 27483 83858 28364 62726 26562 83759 27263 82827 27283
82858 47582 81837 28462 82837 58164 75748 58162 92000
145
BIBLIOGRAPHY
Tritheim (John). Polygraphie et Universelle Ecriture Cabalistique, traduit par Gabriel de Collange. Paris, 1561.
Bright (Timothy), An Arte of Shorte, Swifte and Secret Writing by Character. London, 1588.
Selenus (Gustavus), Cryptomenytics et Cryptographiae. 1624.
England (Parliament of). The Charge of the Common Against Charles Stuart, 1648.
Charles I, Articles Exhibited Against the King, 1648.
Porta (Giovanni Baptista, della). Natural Magick. 1658.
de Vigénère (Blaise), Traite des chiffres. 1670.
Cardano (Girolamo), Aphorismes d'Astrologie de Ptolomie, Hermes, etc. 1657.
Wilkins (John, Bishop of Chester), Mercury, or the Swift Messenger. 1694.
Bacon (Francis), Essays, or Councils, Moral and Civil. 1720.
Thicknesse (Philip). A Treatise on the Art of Deciphering. 1772.
Bacon (Francis), Advancement of Learning. 1772.
Davys (John), An Essay on the Art of Decyphering. 1737.
Astle (Thomas), The Origin and Progress of Writing. 1803.
Hammer (Joseph), Ancient Alphabets and Hieroglyphs Explained. 1806.
Kleuber (j.H.), Kryptographik, L.
Rees (Abraham), W. Blair's Article on 'Cipher' in Cyclopaedia or Universal Dictionary. 1819.
Dalgarno (George), Ars Signorum. 1834.
Vesin de Romanini, La Cryptographie devoilee. 1840.
Vesin de Romanini, Resume de la Cryptographie. 1844.
Morse (E.L), Samuel F. B. Morse, His Letters and Journals.
Morse (S.F.B.), How to Learn Morse Alphabet in Half an Hour. 1876.
Kerckoff (Aug.), La Cryptographie militaire. 1883.
Beaufort, Cryptography, a System of secret Writing. 1882.
Josse (H.), La cryptographie et ses applications a l'art militaire.
Braille (Louis), L'Ecriture a l'usage des aveugles. 1890.
146
Perret (Pail Michel), Les regles de Cicco Simonetta por le dechiffrement de 1474. 1890
Bazeries, E, Chiffres Bazieres. 1893.
Valerio (P.), Cryptographie. 1893.
Schooling (J.H.), Article on Secret Writing in Pall Mall Magazine. 1896.
Bazeries et Bu Gand, La masque de fer. 1893.
Mazzarini (Cardinal), Un agent secret de Mazarin. 1904.
Mansfield (L.C.S.), The Solution of Codes and Ciphers. 1936.
Most people have a natural curiosity to know what lies beyond the closed door, what secrets are hidden behind signs and symbols that have no obvious meaning. When the urge to solve such problems becomes a dominating force in a man’s life, then he may discover new worlds. It may be that there is nothing much left to be discovered in the material world, but there are unlimited possibilities in the world of thoughts and ideas.
d'Agapeyeff
Good Luck !
|
