Qsr Nrwn
Website
Classical Cryptography Course by Lanaki |
Lecture 05Xenocrypt Morphology I |
In Lecture 5, we begin our attack on substitution ciphers created in languages other than English. First, we develop an understanding of cryptography in its role as a cultural universal. Next, we tour the elements of language and the common cryptographic threads that make cryptographic analysis possible. We then look at GERMAN Xenocrypts, applied traffic analysis and the ADFGVX cipher of 1918 WWI vintage.
Xenocrypts are foreign language substitutions. Solving a Xenocrypt (aka XENO) gives double pleasure; not only do you have the fun of solving, but also the satisfaction of knowing that you are acquiring a bowing acquaintance with other languages.
PHOENIX has compiled and edited a Xenocrypt handbook [XEN1] which brings together material published in The Cryptogram since 1940. The book will be available to the KREWE in 1996. It is an excellent tool. Lectures 5-7 will augment his efforts. Quoted from PHOENIX's Preface in reference [XEN1]:
"Don't be afraid of Xenocrypts. The languages used should not offer particular difficulties. Comparing an English printers table (ETAINORSH...) with any of these languages will show a lot of resemblance. That's because English contains elements of most of the languages. Spellings and endings will differ, but there often will be solid 'root' that strongly resembles an English word. Most short English words are of Saxon origin, akin to Danish, Swedish, Dutch, and more remotely German. Longer words come to us from Latin or Norman - French in many instances, and all have cognates in common with English, generally differing slightly from the English version, but often not at all, especially in French. "
In New Orleans, I keynoted the 1994 ACA Convention with the possibility that any language could be learned from its cryptographic building blocks. Xenocrypts represent a cultural universal expressed at its common denominator - mathematics. [NICX]
I suggested that languages be taught in schools first via cryptography and then via sound and structure. This is how I taught myself the rudiments of Russian, Japanese and Korean. Cryptography enhanced my passable understanding of French and reasonable efforts with German.
The real enjoyment came when I could understand Goethe in German, and translated parts of Budo Shoshinshu by the 17 Century author Daidoji Yuzan [SADL]. Solving Xeno's can open our eyes to other cultures.
Linguistic anthropologists have used cryptography to reconstruct ancient languages by comparing contemporary descendants and in so doing make discoveries about history. Others make inferences about universal features of language, linking them to uniformities in the brain. Still others study linguistic differences to discover varied world views and patterns of thought in a multitude of cultures. [KOTT]
The Rossetta Stone found by the Egyptian Dhautpol and the French officer Pierre-François Bouchard near the town of Rosetta in the Nile Delta, gave us a look at Syriac, Greek and Egyptian Hieroglyphs all of the same text. The fascinating story of its decipherment is covered in Kahn. [KAHN] Of special interest was the final decipherment of the Egyptian writing containing homophones - different signs standing for the same sound. [ROSE]
Until the late 1950's linguists thought that the study of language should proceed through a sequence of stages of analysis. The first stage was phonology, the study of sounds used in speech. Phones are speech sounds present and significant in each language. They were recorded using the International Phonetic Alphabet, a series of symbols devised to describe dozens of sounds that occur in different languages.
The next stage was morphology, the study of forms in which sound combine, to form morphemes - words and their meaningful constituents. The word cats has two morphemes /cat/ and /s/ indicating the animal and plurality. A lexicon is a dictionary of all morphemes. A morpheme is the smallest meaningful unit of speech. [MAYA] Isolating or analytic languages are those in which words are morphologically unanalyzable, like Chinese or Vietnamese. Agglutinative languages string together successive morphemes. Turkish is a good example of this. Inflection languages change the form of a word to mark all kinds of grammar distinctions, such as tense or gender. Indo- European languages tend to be highly inflectional.
The next step was to study syntax, the arrangement and order of words in phrases and sentences.
No language contains all the sounds in the International Phonetic Alphabet. Nor is the number of phonemes -significant sound contrasts in a given language - infinite. Phonemes lack meaning in themselves but through sound contrasts distinguish meaning. We find them in minimal pairs, words that resemble each in al but one sound. An example is the minimal pair pit/bit. The /p/ and /b/ are phonemes in English. Another example is bit and beat which separates the phonemes /I/ and /i/ in English. [KOTT] Friedman describes a similar phenomena called homologs and uses them to solve a variety of cryptograms. [FR2] A phoneme is the smallest unit of distinctive sound. [MAYA]
Standard (American) English (SE), the region free dialect of TV network newscasters, has about thirty-five phonemes of at least eleven vowels and twenty four consonants. The number of phonemes varies from language to language - from fifteen to sixty, averaging between thirty and forty. The number of phonemes varies between dialects. In American English, vowel phonemes vary noticeably from dialect to dialect. Readers should pronounce the words in Figure 5-1, paying attention to whether they distinguish each of the vowel sounds. We Americans do not generally pronounce them at all. [BOLI]
Figure 5-1 Vowel Phonemes Standard American English According to Height of Tongue and Tongue Position in Front, Center and Back of Mouth Tongue High
i u
I U
ea ua o
e ou Mid
ae a
Tongue Low
Tongue Central Tongue
Front Back
Phonetic symbols are identified by English words that include them; note that most are minimal pairs.
| high front (spread) | [i] as in beat |
| lower high front (spread) | [I] as in bit |
| mid front (spread) | [ea] as in bait |
| lower mid front (spread) | [e] as in bet |
| low front | [ae] as in bat |
| central | [ua] as in butt |
| low back | [a] as in pot |
| lower mid back (rounded) | [ou] as in bought |
| mid back (rounded) | [o] as in boat |
| lower high back (rounded) | [U] as in put |
| high back (rounded) | [u] as in boot |
Phonetics studies sounds in general, what people actually say in various languages.
Phonemics is concerned with sound contrasts of a particular language. In English /b/ and /v/ are phonemes, occurring in minimal pairs such as bat and vat. In Spanish, the contract between [b] and [v] doesn't distinguish meaning, and are not phonemes. The [b] sound is used in Spanish to pronounce words spelled with either b or v. (Non phonemic phones are enclosed in brackets).
In any language a given phoneme extends over a phonetic range. In English the phoneme /p/ ignores the phonetic contrast between the [pH] in pin and the [p] in spin. How many of you noticed the difference? [pH] is aspirated, so that a puff of air follows the [p]. not true with [p] in spin. To see the difference, light a match and watch the flame as you say the two words. In Chinese the contrast between [p] and [pH] is distinguished only by the contrast between an aspirated and unaspirated [p]. [BOLI]
Norm Chomsky's influential book Syntactic Structures (1957) advocated a new method of linguistic analysis - Transform- ational-generative grammar. [CHOM] Chomsky felt that a language is more than the surface phenomena just discussed (sounds, words, word order). He felt that all languages shared a limed set of organizing principles. Chomsky observed that every normal child who grows up in society develops language easily and automatically. This occurs because the brain contains a genetically transmitted blueprint, or basic linguistic plan for building language. Chomsky called this universal grammar. As children learn their native language, they experiment with their blueprint, reject some sections applying to other languages and gradually focus in and accept the principles of their own language. They do this at about the same age. His study also showed that we learn languages at similar rates. There are universal improper generalizations (foot, foots; hit, hitted) which eventually are corrected.
We master a specific grammar as we learn to speak. These rules let us convert what we want to say into what we do say. People who hear us and speak our language understand our meaning. This works at a cryptographic level also. Chomsky distinguishes between competence (what the speaker must and does know about his language in order to speak and understand) and performance (what a speaker actually says in social situations or writes to someone ). Competence develops during childhood and becomes an unconscious structure. The linguist or cryptographer must discover the structure by looking at deep structures (the mental level) and the surface structure (actual speech) to find the transformational rules that link them. Figure 5-2. shows the Chomsky Model.
Figure 5-2 Chomsky Model For Message From Speaker to Hearer or Writer on Both Sides ... Sounds (phonological component)... . . . . . . Surface-structure sentence Surface-structure sentence . . . . Transformational rule Transformational rule . . . . Deep structure sentence Deep structure sentence . . . . . . Thought Thought (meaning, semantic component (meaning, semantic component ^ SPEAKER HEARERChomsky model tells us why Xenos are so valuable. The human brain contains a limited set of rules for organizing language. The fact that people can learn foreign languages and that words and ideas can be translated from one language into another supports the Chomsky model that all humans have similar linguistic abilities and thought processes.
Other linguists take the view that rather than universal structures as clues to relationships between languages, they belief that different languages produce different thinking and writing. Edward Sapir and Benjamin Whorf argue that speakers think about things in particular ways. For example, the third person singular pronouns of English (he, she, him, her, his, hers) distinguish gender, whereas those of the Palaung of Burma do not. [BURL] [SAPR] [WHOR]
Gender exists in English, although a fully developed noun-gender and adjative-agreement system as in French and other Romance Languages (la belle fille, la beau fils), does not. The Sapir-Whorf hypothesis suggests that English speakers pay more attention to differences between males and females than the Palaung but less than the French and Spanish speakers.
English divides time into past, present, and future. Hopi, a language of the Pueblo region of the Native American Southwest does not. Hopi does distinguish between events that exist or have existed and those don't or don't yet, along with imaginary and hypothetical events. Differing perceptions of time and reality cause difference in spoken and written thought.
A lexicon or vocabulary is a language's dictionary, its set of names for things, events and ideas. APEX DX can probably confirm that Eskimos have several distinct words for snow. In English all forms of snow are the same (unless you are a dope dealer). The Nuer of the Sudan have an elaborate vocabulary to describe cattle. Specialized distinctions between groups is called focal vocabulary. Cattle vocabulary of Texas ranchers is more extensive than New Yorkers; Aspen ski bums differentiate types of snow that are missing from the lexicons of Florida retirees. Ten years ago who would have 'faxed' anything. Simplification of often used words are called monolexemes and compound expressions are simplified such as tropical storm to rain. A television becomes TV, an automobile a car, and a videocassette recorder becomes a VCR.
Semantics refers to a language meaning system. Language, culture and thought are interrelated. There is considerable difference between female and male Americans in regard to color terms. Distinctions implied by such terms as salmon, rust, peach, beige, teal, mauve, cranberry, and dusk orange aren't in the vocabularies of most American men. Ask a fashionable woman and she will know them all. [LAKE]
Knowledge of linguistic relationships is often valuable to determine the events of the past 5000 years. By studying contemporary daughter languages, past language features can be reconstructed. Daughter languages descend from the same parent language that has been changing for thousands of years. The original language from which they diverge is called a protolanguage. French and Spanish are daughter languages of Latin. Language evolves over time into subgroups (closely related from a taxonomy point of view) but with distinct cultural differences. Figure 5-3. shows the main languages and subgroups of the Indo European language stock.
All these daughter languages have developed out of the protolanguage (Proto-Indo-European) spoken in Northern Europe about 5,000 years ago. Note subgroupings. English, a member of the Germanic branch, is more closely related to German and Dutch than it is to Italic or Romance languages such as French and Spanish. However, English shares many linguistic features with French through borrowing and diffusion. [FROM]
The doctrine of linguistic relativity is central to cryptographic treatment of language ciphers. It states that all known languages and dialects are effective means of communication. [KOTT] Nichols Theorem states that if they are linguistically related, they can be codified, enciphered, deciphered and treated as cryptographic units for analysis and statistical treatment. [NICX]
Figure 5 -3 Main Languages of Indo-European Stock INDO-EUROPEAN . ............................................................ . . . . . . . . CELTIC ITALIC GERMANIC . . . . . . . . . . . . . . . o Welsh . . . . o Irish . West North . o Scots Gaelic . . . . o Breton . . . . . . . . ROMANCE o Dutch o Danish . . o English o Icelandic . Latin o Flemish o Norwegian . . o Frisian o Swedish . . o German . o Catalan o Yiddish . o French . o Italian . o Portuguese . o Provencal . o Rumanian . o Spanish . . . ............................................................. . . . . . . . HELLENIC Albanian . . . . . . Armenian . Ancient Greek . . . . . Greek . . . ............................................... . . . . . . INDO-IRANIAN BALTIC SLAVIC . . . . . . . o Latvian o Bulgarian . o Lithuanian o Czech . o Macedonian . o Polish o Old Persian o Russian o Persian o Serbo-Croatian o SANSKRIT o Slovak . o Slovenian . o Ukrainian . o Bengali o Hindi o Punjabi o UrduFigure 5-3 pertains to live languages. Professor Cyrus H. Gordon in his fascinating book "Forgotten Scripts" shows how cryptography is used to recover ancient writings. He tells the story of the unraveling of each of these ancient languages: Egyptian, Old Persion, Sumer-Akkadian, Hittite, Ugaritic, Eteocretan, Minoan and Eblaite. He specializes in cuniform and hieroglyphic inscriptions and gives us a glimpse into the ancient societies that gave birth to the Western world. [GORD] See also references [BARB], [POPE] and [STUR].
There is a common cryptographic thread for most languages. All known writing systems are partly or wholly phonetic, and express the sounds of a particular language. Writing is speech put in visible form, in such a way that any reader instructed in its conventions can reconstruct the vocal message. Writing as "visible speech" was invented about five thousand years ago by Sumerians and almost simultaneously by ancient Egyptians.
The ancient Mayan knew that it was 12 cycles, 18 katuns, 16 tuns, 0 uinals, and 16 kins since the beginning of the Great Cycle. The day was 12 Cib 14 Uo and was ruled by the seventh Lord of the Night. The moon was nine days old. Precisely 5,101 of our years and 235 days had passed. So said the ancient Mayan scribes. We remember the day as 14 May 1989.
Three kinds of writing systems have been identified: Rebus which is a combination of logograms and phonetic signs; Syllabic such as CV - consonant vowel such as Cherokee or Inuit; and Alphabetic, which is phonemic, the individual consonants and vowels make up the sounds of the language.
Table 5-2 differentiates writing systems by the number of signs used. [MAYA]
| Writing System | No. of Signs |
| Logographic | |
| Sumerian | 600+ |
| Egyptian | 2,500 |
| Hittite Hieroglyphic | 497 |
| Chinese | 5,000+ |
| "Pure" Syllabic | |
| Persian | 40 |
| Linear B | 87 |
| Cypriote | 56 |
| Cherokee | 85 |
| Alphabetic or Consonantal | |
| English | 26 |
| Anglo-Saxon | 31 |
| Sanskrit | 35 |
| Etruscan | 20 |
| Russian | 36 |
| Hebrew | 22 |
| Arabic | 28 |
Michael D. Coe classifies the entire Proto-Mayan languages. In fourteen daughter divisions of Proto-Mayan, there are thirty one sub languages from Huastec to Tzuthil. Extraordinary story of applied cryptanalysis and applied linguistics. [MAYA]
I used to think that Xenocrypts - non English cryptograms, were very difficult to solve. The 'aha' light came on several years ago, when I realized that most languages share the common framework of mathematics and statistics. To be able to solve Xenocrypts, it is only necessary to learn the basic (group) mathematical structure of the language, to use a bidirectional translation dictionary and to recognize the underlying cipher construct. [NICX]
Many challenge ciphers start with the problem of recognizing the language and then the distribution of characters within the particular language. The legendary W. F. Friedman once remarked: "treating the frequency distribution as a statistical curve, when such treatment is possible, is one of the most useful and trustworthy methods in cryptography." [FR1], [FRE]
Table 1 gives the frequency distributions of ten of my favorite languages (sans Russian, Chinese and Japanese which require character sets that will not transfer via my e-mail). The frequencies in Table 5-1 have been developed from various sources. Table 5-1 frequencies may differ from other published data, based on text derived solely from literature or military sources, because I have included the practical text from my solved Xeno's over the years. Letters used in cryptograms tend to shift the frequency distribution. Frequencies of letters, and their order, are not fixed quantities in any language. Group frequencies, however, are fairly constant in every language. This is the common thread - the linguistic relativity of all languages. [NICX], [NIC1]
TABLE 5-1 Partial Frequency Distribution For Cracking Xenocrypts 16 8 7 6 5 4 2 <1 NORWEGIAN: E RNS T AI LDO GKM UVFHPA' JBO' YAECWXZQ 10 9 7 6 4 3 <2 LATIN: I E UTA SRN OM CPL (bal) 18 8 7 6 5 4 3 2 <1 FRENCH: E AN RSIT UO L D CMP VB F-Y 14 13 12 8 6 5 4 3 2 <1 PORTUGUESE: A E O RS IN DMT UCL P QV (bal) 18 11 8 7 5 4 3 2 <1 GERMAN: E N I RS ADTU GHO LBM CW (bal) 15 12 8 7 5 4 3 1 <1 CATALAN: E A S ILRNT OC DU MP BVQGF (bal) 16 13 8 6 5 4 3 <2 HUNGARIAN: E A T OS LNZ KIM RGU (bal) 13 12 11 9 7 6 5 3 2 <1 ITALIAN: E A I O L NRT SC DMO'U VG (bal) 20 10 7 6 5 4 3 2 <1 DUTCH: E N IAT O DL S GKH UVWBJMPZ (bal) 13 9 8 7 5 4 3 1 <1 SPANISH: EA O S RNI DL CTU MP GYB (bal)English has its characteristic frequencies and sequence data (based on 10,000 letters):
% 12 10 8 8 7 7 7 6 5 4-3 2 1 < 1 ENGLISH: E / T A / O N I S R H / LDCU / PFMW / YBGV / KQXJZ GROUP PERCENTAGES: A E I O U 38.58% L N R S T 33.43% J K Q X Z 1.11% E T A O N 45.08% E T A O N I S R H 70.02% ORDER Digram Order: TH / HE / AN / IN / ER / RE / ES / ON / EA / TI / AT / ST / EN / ND / OR Trigram Order: THE / AND / THA / ENT / ION / TIO / FOR / NDE Reversals: ER RE / ES SE / AN NA /TI IT /ON NO / IN NI Initials: T A O S H I W C B P F D M R Finals: E S T D N R O Y Vowel % 40% (y included)The ACA Xenocrypt Handbook compiled by PHOENIX, develops similar mathematical data on fifteen languages presented in The Cryptogram on a regular basis. [XEN1]
Review Lecture 2 Kullback's tests and Friedman's I.C. test.
Kullback gives the following tables for Monoalphabetic and Digraphic texts for eight languages:
Note that the English plain text value is slightly less than Friedman's. [KULL] [SINK]
| Monographic Text | Digraphic Test | |
| English | 0.0661 N (N - 1) | 0.0069 N (N - 1) |
| French | 0.0778 N (N - 1) | 0.0093 N (N - 1) |
| German | 0.0762 N (N - 1) | 0.0112 N (N - 1) |
| Italian | 0.0738 N (N - 1) | 0.0081 N (N - 1) |
| Japanese | 0.0819 N (N - 1) | 0.0116 N (N - 1) |
| Portuguese | 0.0791 N (N - 1) | |
| Russian | 0.0529 N (N - 1) | 0.0058 N (N - 1) |
| Spanish | 0.0775 N (N - 1) | 0.0093 N (N - 1) |
| Monographic | Digraphic | Trigraphic |
| 0.038 N (N - 1) | 0.0015 N (N - 1) | 0.000057 N (N - 1) |
XENO's - foreign language substitutions, as given in the Xenocrypt Department of The Cryptogram, are usually quotations, or simple normal wording. Thus the Frequency Table of a Xenocrypt will follow closely to the normal Frequency Table of its language. Arranging these two tables in order of frequency, rather than alphabetically, may be used for testing probable equivalents. When words are found, if the meaning is not known, a dictionary helps.
The Contact and Position Tables are used just as in solving English cryptograms.
Lets start off with German Xenocrypts.
| E | 10,778 |
| N | 6,336 |
| I | 4,879 |
| R | 4,339 |
| S | 4,127 |
| A | 3,601 |
| T | 3,447 |
| D | 3,248 |
| U | 2,753 |
| H | 2,477 |
| L | 1,988 |
| G | 1,921 |
| O | 1,635 |
| C | 1,620 |
| M | 1,360 |
| B | 1,023 |
| F | 958 |
| W | 899 |
| K | 747 |
| Z | 654 |
| V | 523 |
| P | 499 |
| J | 192 |
| Y | 24 |
| X | 12 |
| Q | 6 |
| 60,046 |
Monographic Kappa Plain, German Language = 0.0787, I.C. = 2.05
| E | 180 |
| N | 106 |
| I | 81 |
| R | 72 |
| S | 69 |
| A | 60 |
| T | 57 |
| D | 54 |
| U | 46 |
| H | 41 |
| L | 33 |
| G | 32 |
| O | 27 |
| C | 27 |
| M | 23 |
| B | 17 |
| F | 16 |
| W | 15 |
| K | 13 |
| Z | 11 |
| V | 9 |
| P | 8 |
| J | 3 |
| Q | - |
| X | - |
| Y | - |
| 1,000 |
| D | 1,716 |
| A | 762 |
| S | 698 |
| E | 686 |
| I | 581 |
| U | 550 |
| W | 544 |
| G | 461 |
| B | 460 |
| V | 408 |
| Z | 343 |
| M | 339 |
| N | 306 |
| F | 280 |
| H | 265 |
| K | 263 |
| P | 181 |
| R | 167 |
| L | 158 |
| O | 135 |
| J | 135 |
| T | 106 |
| C | 22 |
| Q | 2 |
| 9,568 |
| 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 | |
| A | 4 | 14 | 10 | 4 | 33 | 7 | 9 | 7 | 1 | 1 | 2 | 33 | 13 | 48 | 2 | 22 | 27 | 23 | 36 | 1 | 1 | 1 | ||||
| B | 6 | 48 | 1 | 1 | 5 | 3 | 3 | 11 | 2 | 1 | 3 | 1 | 1 | |||||||||||||
| C | 130 | 5 | ||||||||||||||||||||||||
| D | 29 | 2 | 8 | 127 | 1 | 2 | 2 | 60 | 1 | 3 | 2 | 2 | 4 | 1 | 5 | 6 | 2 | 9 | 2 | 2 | 2 | |||||
| E | 13 | 22 | 10 | 31 | 13 | 12 | 32 | 24 | 90 | 2 | 6 | 28 | 25 | 235 | 3 | 6 | 195 | 68 | 28 | 24 | 9 | 15 | 7 | |||
| F | 7 | 1 | 3 | 15 | 7 | 2 | 2 | 2 | 1 | 1 | 3 | 10 | 2 | 10 | 12 | |||||||||||
| G | 10 | 1 | 8 | 78 | 1 | 2 | 2 | 8 | 2 | 7 | 1 | 3 | 1 | 11 | 8 | 5 | 8 | 2 | 1 | 1 | ||||||
| H | 29 | 1 | 8 | 64 | 1 | 2 | 1 | 14 | 2 | 8 | 3 | 6 | 6 | 1 | 20 | 4 | 23 | 7 | 2 | 3 | 1 | |||||
| I | 3 | 1 | 39 | 7 | 91 | 2 | 18 | 7 | 2 | 7 | 12 | 11 | 84 | 13 | 1 | 7 | 53 | 44 | 1 | 2 | 1 | 1 | ||||
| J | 4 | 8 | 3 | |||||||||||||||||||||||
| K | 12 | 1 | 1 | 11 | 1 | 1 | 1 | 5 | 9 | 10 | 1 | 5 | 4 | |||||||||||||
| L | 26 | 3 | 1 | 6 | 27 | 1 | 2 | 37 | 3 | 20 | 1 | 2 | 4 | 10 | 12 | 6 | 1 | 1 | ||||||||
| M | 16 | 3 | 4 | 26 | 2 | 22 | 1 | 14 | 1 | 2 | 1 | 11 | 1 | 8 | 5 | 1 | 3 | 3 | 9 | 1 | 1 | 1 | ||||
| N | 39 | 12 | 118 | 58 | 9 | 57 | 8 | 35 | 4 | 10 | 6 | 10 | 18 | 18 | 8 | 5 | 4 | 36 | 27 | 20 | 10 | 17 | 14 | |||
| O | 1 | 3 | 5 | 3 | 11 | 3 | 3 | 3 | 1 | 18 | 6 | 33 | 1 | 5 | 18 | 12 | 4 | 1 | 1 | 5 | 1 | |||||
| P | 10 | 5 | 4 | 1 | 2 | 1 | 7 | 2 | 7 | 1 | 1 | |||||||||||||||
| Q | 1 | |||||||||||||||||||||||||
| R | 34 | 11 | 5 | 35 | 60 | 9 | 12 | 9 | 37 | 2 | 11 | 6 | 8 | 12 | 19 | 3 | 6 | 22 | 18 | 26 | 6 | 8 | 5 | |||
| S | 14 | 6 | 55 | 13 | 46 | 3 | 7 | 3 | 30 | 1 | 5 | 4 | 7 | 3 | 16 | 6 | 2 | 40 | 57 | 9 | 5 | 5 | 1 | 5 | ||
| T | 25 | 3 | 17 | 88 | 2 | 4 | 6 | 40 | 1 | 3 | 7 | 3 | 4 | 4 | 14 | 20 | 7 | 16 | 2 | 10 | 13 | |||||
| U | 1 | 2 | 8 | 2 | 37 | 15 | 5 | 1 | 2 | 2 | 11 | 76 | 2 | 18 | 28 | 14 | 1 | 1 | 2 | 1 | ||||||
| V | 1 | 19 | 3 | 21 | ||||||||||||||||||||||
| W | 16 | 24 | 20 | 3 | 6 | 6 | ||||||||||||||||||||
| X | ||||||||||||||||||||||||||
| Y | ||||||||||||||||||||||||||
| Z | 1 | 1 | 8 | 5 | 1 | 2 | 4 | 27 | 4 |
Digraphic Kappa plain = 0.0111, I.C. = 7.50
| EN | 235 |
| ER | 195 |
| CH | 130 |
| DE | 127 |
| ND | 118 |
| IE | 91 |
| EI | 90 |
| TE | 88 |
| IN | 84 |
| GE | 781,236, a. |
| UN | 76 |
| ES | 68 |
| HE | 64 |
| DI | 60 |
| RE | 60 |
| NE | 58 |
| NG | 57 |
| ST | 57 |
| SC | 55 |
| IS | 53 |
| AN | 48 |
| BE | 48 |
| SE | 46 |
| IT | 44 |
| SS | 40 |
| TI | 40 |
| IC | 39 |
| NA | 39 |
| LI | 37 |
| RI | 37 |
| UE | 37 |
| AU | 36 |
| NS | 36 |
| NI | 35 |
| RD | 35 |
| RA | 34 |
| AE | 332,508, b. |
| AL | 33 |
| ON | 33 |
| EG | 32 |
| ED | 31 |
| SI | 30 |
| DA | 29 |
| HA | 29 |
| EL | 28 |
| ET | 28 |
| US | 28 |
| AS | 27 |
| LE | 27 |
| NT | 27 |
| ZU | 27 |
| LA | 26 |
| ME | 26 |
| RU | 26 |
| EM | 25 |
| TA | 25 |
| EH | 24 |
| EU | 24 |
| WE | 24 |
| AT | 23 |
| HT | 23 |
| AR | 22 |
| EB | 22 |
| RS | 22 |
| VO | 21 |
| HR | 20 |
| LL | 20 |
| NU | 20 |
| TS | 20 |
| WI | 20 |
| RO | 19 |
| VE | 19 |
| NN | 18 |
| OL | 18 |
| OR | 18 |
| RT | 18 |
| UR | 18 |
| IG | 17 |
| NW | 17 |
| MA | 16 |
| SO | 16 |
| TD | 16 |
| TU | 16 |
| WA | 16 |
| FE | 15 |
| UF | 15 |
| AB | 14 |
| EW | 14 |
| HI | 14 |
| MI | 14 |
| NZ | 14 |
| SA | 14 |
| TR | 14 |
| UD | 14 |
| SD | 13 |
| 3,747 |
| EN | 235 |
| ER | 195 |
| DE | 127 |
| IE | 91 |
| EI | 90 |
| IN | 84 |
| GE | 78 |
| ES | 68 |
| RE | 60 |
| NE | 58 |
| IS | 53 |
| AN | 48 |
| SE | 46 |
| IT | 44 |
| TI | 40 |
| NA | 39 |
| NI | 35 |
| EG | 32 |
| ED | 31 |
| SI | 30 |
| CH | 130 |
| HC | 0 |
| ND | 113 |
| DN | 2 |
| NG | 57 |
| GN | 3 |
| SC | 55 |
| CS | 0 |
| SS | 40 | EE | 13 | FF | 7 | RR | 6 | GG | 2 | PP | 2 | OO | 1 |
| LL | 20 | MM | 11 | TT | 7 | AA | 4 | II | 2 | HH | 1 | UU | 1 |
| NN | 18 | DD | 8 |
| DE | 805 | EI | 300 | DA | 244 | WE | 192 | ER | 153 | ZU | 124 | ST | 112 |
| DI | 567 | GE | 299 | VO | 214 | VE | 172 | HA | 140 | MI | 117 | IN | 111 |
| UN | 428 | BE | 252 | SI | 197 | WI | 155 | AL | 134 | SN | 112 | SE | 111 |
| AU | 318 |
| SCH | 666 | ERE | 313 | NEN | 198 | AUS | 162 | IST | 142 | HRE | 124 |
| DER | 602 | ENS | 270 | SSE | 191 | TIS | 159 | STA | 141 | HER | 122 |
| CHE | 599 | CHT | 264 | REI | 190 | BER | 157 | DES | 140 | ACH | 119 |
| DIE | 564 | NGE | 263 | TER | 188 | ENI | 157 | FUE | 139 | GES | 118 |
| NDE | 541 | NDI | 259 | REN | 185 | ENG | 155 | NTE | 139 | ABE | 117 |
| EIN | 519 | IND | 254 | EIT | 184 | ION | 154 | UER | 138 | ERA | 117 |
| END | 481 | ERD | 248 | EBE | 178 | SEN | 152 | ERU | 137 | BEN | 116 |
| DEN | 457 | INE | 247 | ENE | 175 | ITI | 151 | TUN | 136 | MEN | 115 |
| ICH | 453 | AND | 246 | LIC | 175 | AUF | 149 | SEI | 133 | RIE | 112 |
| TEN | 425 | RDE | 239 | EGE | 173 | IES | 149 | ESE | 132 | VER | 110 |
| UNG | 377 | ENA | 214 | DAS | 172 | ASS | 148 | ERT | 128 | LAN | 109 |
| HEN | 332 | ERS | 212 | ENU | 171 | ENW | 148 | NDA | 127 | ENB | 108 |
| UND | 331 | EDE | 209 | NUN | 169 | ENT | 146 | IED | 126 | ESS | 108 |
| GEN | 321 | STE | 205 | NER | 166 | ERI | 143 | ERN | 125 | LLE | 108 |
| ISC | 317 | VER | 204 | RUN | 163 | EST | 142 | NAU | 108 | TSC | 107 |
| ENN | 106 | ERG | 106 | RIT | 106 | EHR | 105 | CHA | 104 | VON | 104 |
| SIC | 103 | IGE | 102 | ITE | 101 | ENZ | 100 | ERB | 100 | EUT | 100 |
| EIN | 242 | DAS | 79 | SCH | 73 | AUF | 64 | DEU | 61 | UNT | 57 |
| VER | 170 | BRI | 79 | AUS | 69 | NER | 63 | GES | 60 | GRO | 56 |
| FUE | 89 | DIE | 76 | SEI | 68 | IND | 62 | GEG | 59 | AUC | 55 |
| SIC | 86 | NIC | 73 | STA | 65 | ALL | 61 | UEB | 53 | POL | 52 |
| WIR | 51 |
| SCHE | 398 | NUND | 106 | NICH | 80 | ATIO | 65 | RSCH | 60 | ENZU | 54 |
| ISCH | 317 | ITIS | 104 | UNGD | 80 | GEND | 65 | EDEN | 59 | ITEN | 54 |
| CHEN | 296 | SICH | 103 | EITE | 79 | TEND | 65 | ERGE | 59 | KRIE | 54 |
| NDER | 243 | RUNG | 101 | DEUT | 78 | EBER | 67 | ESSE | 59 | RIEG | 54 |
| EINE | 218 | ANDE | 100 | FUER | 78 | GEGE | 65 | UNTE | 59 | SDIE | 54 |
| ENDE | 216 | UNGE | 100 | CHTE | 77 | POLI | 64 | EICH | 58 | URCH | 53 |
| NDIE | 176 | EREI | 94 | EGEN | 76 | SIND | 64 | TLIC | 58 | ALLE | 52 |
| LICH | 168 | TION | 93 | NEIN | 76 | TUNG | 64 | INER | 57 | DERS | 52 |
| ICHT | 151 | SEIN | 92 | IESE | 75 | ENSI | 64 | EBEN | 56 | ENWE | 52 |
| TISC | 146 | IEDE | 91 | ERST | 74 | FUTS | 64 | ENDA | 56 | HABE | 52 |
| ERDE | 144 | LAND | 91 | RDIE | 74 | LITI | 62 | ENST | 56 | ONEN | 52 |
| ENDI | 141 | SSEN | 90 | ERDI | 72 | UEBE | 62 | IGEN | 56 | SCHI | 52 |
| NDEN | 136 | BRIT | 89 | STEN | 72 | UTSC | 62 | ONDE | 56 | DEND | 51 |
| RDEN | 133 | DASS | 86 | CHER | 71 | AUCH | 62 | TENS | 56 | DISC | 51 |
| ENUN | 120 | NTER | 86 | INDI | 71 | DENS | 62 | EDIE | 55 | ENEN | 51 |
| ICHE | 120 | EDER | 83 | REIN | 71 | EIND | 61 | ERTE | 55 | NACH | 51 |
| INDE | 111 | EREN | 83 | DERE | 70 | OLIT | 61 | HREN | 55 | NDAS | 51 |
| NGEN | 110 | ENGE | 81 | NGDE | 70 | SCHA | 61 | TDIE | 55 | UNGS | 51 |
| ERUN | 109 | ENAU | 80 | ENBE | 68 | SCHL | 61 | ATEN | 55 | ABEN | 50 |
| DIES | 108 | ENIN | 80 | RITI | 66 | WERD | 61 | DIEB | 54 | NBER | 50 |
| TSCH | 107 |
|
|
GD QSMJ TE GSK EVGHSIEKSDNRGK-OGFJDNRGH EVEJGFH HFKOPFKI KGJL SV VSJJGUAGDJUSNRG DJEEJGK EV *Z. *D. EUUGK PFKIGHK DXHGNRGK MGSOG GKQUSDNR FKO OGFJDNR.
A frequency analysis of Ger-1 yields:
G - 20 16.1% Try G=e. K - 13 10.5% Try K=n. J - 10 8.1% Try J=i. S - 9 7.3% D,E - 9 7.3% F - 7 5.6% N,R,H - 6 4.8% V,O,U - 5 4.0% I - 3 P,Q,M - 2 X,Z,A,T,L - 1 B,C,W,Y - 0
1 2 3 4 5 6
e i ein e i ni en e e e
GD QSMJ TE GSK EVGHSIEKSDNRGK-OGFJDNRGH EVEJGFH
7 8 9 10 11
n n ne i i e e i e en
HFKOPFKI KGJL SV VSJJGUAGDJUSNRG DJEEJGK EV
12 13 14 15 16 17 18 19
en e n e en gi e en i n
Z. D. EUUGK PFKIGHK DXHGNRGK MGSOG GKQUSDNR FKO
20
e
OGFJDNR.
So the first three letters follow the German frequency table. Note we have ein. Word 19 is und? and word 1 might be es. The frequencies match. Try these substitutions.
1 2 3 4 5 6 es i ein e i nis en deu s e eu GD QSMJ TE GSK EVGHSIEKSDNRGK-OGFJDNRGH EVEJGFH 7 8 9 10 11 und n ne i i e es i e s en HFKOPFKI KGJL SV VSJJGUAGDJUSNRG DJEEJGK EV 12 13 14 15 16 17 18 19 u s en u e n s e en eide en i s und *Z. *D. EUUGK PFKIGHK DXHGNRGK MGSOG GKQUSDNR FKO 20 deu s OGFJDNR.
A common trigram is sch. Word 20 might be deutsch. Word 1 could be es followed by gibt. Word 17 might be beide.
1 2 3 4 5 6 es gibt ein e i nischen deutscher teur GD QSMJ TE GSK EVGHSIEKSDNRGK-OGFJDNRGH EVEJGFH 7 8 9 10 11 rund n net i ittel estlic e st ten HFKOPFKI KGJL SV VSJJGUAGDJUSNRG DJEEJGK EV 12 13 14 15 16 17 18 19 u s en un e n sprechen beide englisch und *Z. *D. EUUGK PFKIGHK DXHGNRGK MGSOG GKQUSDNR FKO 20 deutsch OGFJDNR.
Word 18 becomes english and word 16 could be speaks in german = sprechen. (insert above)
I note that I have missed a high frequency letter pair E=a. Inserting brings three additional words.
1 2 3 4 5 6 es gibt a ein americanischen-deutscher amateur GD QSMJ TE GSK EVGHSIEKSDNRGK-OGFJDNRGH EVEJGFH 7 8 9 10 11 rund n net im mittelwestliche staaten am HFKOPFKI KGJL SV VSJJGUAGDJUSNRG DJEEJGK EV 12 13 14 15 16 17 18 19 u s allen un e n sprechen beide englisch und *Z. *D. EUUGK PFKIGHK DXHGNRGK MGSOG GKQUSDNR FKO 20 deutsch OGFJDNR.
The flow of the german now is clear. A little worterbuch gives us the balance of letter relationships.
1 2 3 4 5 6 es gibt ja ein americanischen-deutscher amateur GD QSMJ TE GSK EVGHSIEKSDNRGK-OGFJDNRGH EVEJGFH 7 8 9 10 11 rundfunk netz im mittelwestliche staaten am HFKOPFKI KGJL SV VSJJGUAGDJUSNRG DJEEJGK EV 12 13 14 15 16 17 18 19 u s allen funkern sprechen beide englisch und *Z. *D. EUUGK PFKIGHK DXHGNRGK MGSOG GKQUSDNR FKO 20 deutsch OGFJDNR.
The keyword = sauerkraut.
Note the simularities to English Aristocrat solving and to English endings and words. Note the group statistics of the two languages and my comments on common threads. Do you see how this commonality flows from Figure 5-1?
Lets remove the word divisions and try a German Patristocrat.
Ger-2. Traurige Wahrheit. (zwei ewige) Eng K4 GEMINATOR 1 2 3 4 5 6 7
JGKMH FDZJM JZMKJ IMRKJ ICGXR MYJWG XQXRI
8 9 10 11 12 13 14
IMJQJ RGELP MELJI XQQLJ MFCHJ WQMFI JQXRM
15 16 17 18 19 20 21
YJWGX QMGFI CGRME LFKCR DGMEL JWCPH JWFJM
22 23
RGFJM R.
The hint tells us that the words [zwei ewige] is in the cryptogram plain text. We also know that K4 password scheme has been used. Nichols rule says ignore the descriptive part in the title as a red hering.
Start with the frequency analysis:
J - 17 15.3% K - 5 4.5% O - 0 M - 15 13.5% C - 5 4.5% A - 0 R - 9 8.1% W - 5 4.5% B - 0 G - 9 8.1% E - 4 3.6% N - 0 I - 7 6.3% H - 3 2.7% T - 0 Q - 7 6.3% Z - 2 1.8% S - 0 X - 6 5.4% Y - 2 1.8% V - 0 F - 6 5.4% P - 2 1.8% U - 0 L - 5 4.5% D - 2 1.8%
Let J=e and note the patterns at groups 2 and 3 for the hint zwei ewige. So Z=w, D=z, M=i K=g.
1 2 3 4 5 6 7
e gi zwei ewige i ge i e
JGKMH FDZJM JZMKJ IMRKJ ICGXR MYJWG XQXRI
8 9 10 11 12 13 14
ie e i e e i e i e i
IMJQJ RGELP MELJI XQQLJ MFCHJ WQMFI JQXRM
15 16 17 18 19 20 21
e i i g z i e e ei
YJWGX QMGFI CGRME LFKCR DGMEL JWCPH JWFJM
22 23
ei
RGFJM R.
The G is a high frequency letter and could be S, A, or N. Try 'es gibt' in groups 1 and 2. s works, b works, t might.
1 2 3 4 5 6 7 esgib tzwei ewige i ge s i e s JGKMH FDZJM JZMKJ IMRKJ ICGXR MYJWG XQXRI 8 9 10 11 12 13 14 ie e s i e e it be it e i IMJQJ RGELP MELJI XQQLJ MFCHJ WQMFI JQXRM 15 16 17 18 19 20 21 e s i t s i tg z i e b e tei YJWGX QMGFI CGRME LFKCR DGMEL JWCPH JWFJM 22 23 stei RGFJM R.
Now we must find the n, r and the a. R might be our n. (see last group). And QQ = mm, A long leap for C=a by frequency only - later to confirm by digrams. A short leap lets us assume W=r. Placing these guesses in temporarily, we find the following:
1 2 3 4 5 6 7 esgib tzwei ewige dinge dasun ivers umund JGKMH FDZJM JZMKJ IMRKJ ICGXR MYJWG XQXRI 8 9 10 11 12 13 14 dieme nschl iched ummhe itabe rmitd emuni IMJQJ RGELP MELJI XQQLJ MFCHJ WQMFI JQXRM 15 16 17 18 19 20 21 versu mistd asnic htgan zsich eralb ertei YJWGX QMGFI CGRME LFKCR DGMEL JWCPH JWFJM 22 23 nstei n RGFJM R.
Our digram table helps us with cipher text L and X. X is a good candidate for u and L = h is a reasonable guess, because EL = ch brings us two words. Note group 12 now gives us the W=r and I = d! A little help from the dictionary yields Y=v and P=l.
Putting the word divisions back in we have a quote by Dr. Einstein.
Es gibt zwei ewige dinge das universum und die menschliche dummheit aber mit dem universum ist das nicht ganz sicher. == Albert Einstein.
The kewords are (facts; SAD). The plain text x is over the cipher text S for the initial position of the keying alphabets.
A small sister to cryptanalysis is the applications of traffic analysis. Traffic analysis was the forerunner to differential cryptanalysis and a primary reason for the cracking of the German Codes in WWII. {Unfortunately, the same principles worked on the British and American Codes as well.} The German Army (maybe even the German Soul) was dedicated to unquestioned organization. Paperwork and radio messages must flow to the various military units in a prescribed manner. Traffic Analysis is the branch of signal intelligence analysis which deals with the study of external characteristic of signal communications.
The information is used:
Allowing for differences in language and procedure signs and signals, there are six standard elements for military radio communications systems. These are:
In order to insure proper handling of messages in the field and message center, some information was sent in the clear or using simple coding. This information about routing and accounting was usually in the preamble or message postamble. This included:
As a general rule, German high-echelon traffic contained most of these items and German low-echelon traffic cut them to a minimum.
The German penchant for organization could be seen in the way they handled serial numbers. Any radio message flowing from division level to soldier in the field would have a reference serial number attached in clear or matrix cipher, by the writer, the HQ message center, the signal center or code room, the "in desk" , the transmitter, linkage, and/or operator. The routing system usually consisted of a code and syllabary that represented the location or unit. [HIN1]
An example taken from WWII U. S. Army procedure:
A45 BR6 B STX-O-P P-A45 BR6-T-N-A45 A-79K 011046Z A-45-W-F2P SLW BR6 GR 28 BT TEXT BT 011046Z K
where:
A45 BR6 - multiple callup; receiving calls STX-O-P - transmitting call with precedence designation, OP = operational priority P-A45 - message priority to A45 only; to others routine BR6-T-N-A45 - BR6 to relay to all except A45 A-79K - originator of message 011046 - Date and Time Zulu used pre and postamble A-45 - action destination W-F2P SLW BR6 - Information destinations GR 28 Group Count.. note how small for such external information envelope
You can see where modern E-Mail and word processing systems have made some of this information easier to handle by the portable desk idea but traffic analysis would still apply.
American "cryptees" were adept in determining the German Order of Battle from their cryptonets (ex. from intercepts re limited distribution from corp to a theater). Traffic analysis not only gave the locations but the communication relationships between units or groups of units in the field. Some German commands were allowed latitude in their compositions of codes and ciphers. This proved to be an exploitable fault in the German security.
American success in reconstructing German communication networks was partly do to the appropriate (and sometimes lucky) analysis of the routing system. The radio station could be tied into the code group. Crib techniques included focusing on the relay point, recognizing a book message crib to several locations, correlating the address and signature cribs, tagging the operational chatter, separating the addresses, using solved messages to give outright routing assignments, syllabary solutions and changes in the system itself.
The textual features of the message gave valuable information. Tabulations of messages, text type, and volumes helped discriminate the practice and dummy traffic. Recognition of the communications net as order of battle often gave away the crypto-entity.
Traffic analysis yields information via Crib messages, Isologs and Chatter. Crib messages assume a partial knowledge of the underlying plain text through recognition of the external characteristics. Command sitrep reports, up and down German channels, were especially easy for American crypees. The origin, serial number range, the cryptonet id, report type, the file date and time, message length and error messages in the clear, gave a clear picture of the German command process. German order of battle, troop dispositions and movements were deduced by traffic analysis. [TM32]
An Isolog exists when the underlying plain text is encrypted in two different systems. They exist because of relay repetition requirements, book messages to multiple receivers (spamming would have been a definite no-no), or error by the code clerk. American crypees were particularly effective in obtaining intelligence from this method.
Traffic analysis boils down to finding the contact relationships among units, tracking their movements, building up the cryptonet authorities, capitalizing on lack of randomness in their structures, and exploiting book and relay cribs. I submit that American intelligence was quite successful in this endeavor against the Germans in WWII.
Jaeger was a German code expert sent to stiffen the German Code discipline in France in 1918. Ironically, the double "e" in Jaeger's name gave US Army traffic analysis experts a fix on code changes in 1918.
ADFGVX, is one of the best known field ciphers in the history of cryptology. Originally a 5 x 5 matrix of just 5 letters, ADFGX, the system was expanded on June 1, 1918 to a 6th letter V. The letters were chosen for their clarity in Morse: A .-, D -.., F ..-., G --., V ...-, and X -..-.
W. F. Friedman describes one of the first traffic analysis charts regarding battle activity from May to August, 1918 at Marne, and Rheims, France. It was based solely on the ebb and flow of traffic in the ADFGVX cipher. This cipher was restricted to German High Command communications between and among the headquarters of divisions and army corps.
The ADFGVX cipher was considered secure because it combined both a good substitution (bipartite fractionation) and an excellent transposition in one system. During the eight month history of this cipher, only 10 keys were recovered by the Allies (in 10 days of heavy traffic) and fifty percent of the messages on these days were read. These intercepts effected the reverse of the German advances (15 divisions) under Ludendorff at Montdidier and Compiegne, about 50 miles North of Paris. Solution by the famed French Captain Georges Painvin was based on just two specialized cases. No general solution for the cipher was found by the Allies. In 1933, William Friedman and the SIS found a general solution. French General Givierge, of the Deuxieme Bureau also published a solution to the general case.
The June 3 message that Painvin cracked which changed the course of WWI:
From German High Command in Remaugies:
Munitionierung beschleunigen Punkt Soweit nicut eingesehen auch bei TagCT starts: CHI-126: FGAXA XAXFF FAFFA AVDFA GAXFX FAAAG
This told the Allies where and when the bombardment preceding the next major German push was planned.
26 letters and 10 digits of the ADFGVX were placed into a 6 x 6 Bipartite Square:
| A | D | F | G | V | X | |
| A | F | L | 1 | A | O | 2 |
| D | J | D | W | 3 | G | U |
| F | C | I | Y | B | 4 | P |
| G | R | 5 | Q | 8 | V | E |
| V | 6 | K | 7 | Z | M | X |
| X | S | N | H | 0 | T | 9 |
| PT | a | l | l | q | u | i | e | t | o | n | t | h | i | s | f | r | o | n | t | t | o | d | a | y | |||||
| CT | AG | AD | AD | GF | DX | FD | GX | XV | AV | XD | XV | XF | FD | XA | AA | GA | AV | XD | XV | XV | AV | DD | AG | FF |
The bilateral cipher which results is transposed with a keyed matrix, written in by row and removed by column.
| G | E | R | M | A | N |
| 3 | 2 | 6 | 4 | 1 | 5 |
| A | G | A | D | A | D |
| G | F | D | X | F | D |
| G | X | X | V | A | V |
| X | D | X | V | X | F |
| F | D | X | A | A | A |
| G | A | A | V | X | D |
| X | V | X | V | A | V |
| D | D | A | G | F | F |
and the final CT is:
AFAXA XAFGF XDDAV DAGGX FGXDD XVVAV VGDDV FADVF ADXXX AXA
Known decipherment was accomplished with the Key and possession of the original matrix. Fine and dandy but cryptanalysis in 1918, was another thing.
According to William Friedman, there were only three viable ways to attack this cipher. The first method required 2 or messages with identical plain text beginnings to uncover the transposition. Under the second method, 2 or more messages with plain text endings were required to break the flat distribution shield of the substitution part of the cipher. The German addiction to stereotyped phraseology was so prevalent in all German military communications that in each days traffic, messages with similar endings and beginnings were found (sometimes both). The third method required messages with the exact same number of letters. Painvin used the first two methods when he cracked the 5 letter ADFGX version in April, 1918. [FRAA], [FRAB]
Lest we underestimate the difficulty of this cipher, I think we might step behind Painvin shoulders as he worked. At 4:30 am on March 21, 6000 guns opened fire on the Allied line at Somme. Five hours later, 62 German Divisions pushed forward on a 40 mile front. Radio traffic increased dramatically, Painvin had just a few intercepts in the ADFGX cipher and the longer ones had been split in three parts to prevent anagraming.
Five letters, therefore, a checkerboard? Simple mono cipher - too flat a distribution.
The German oddity of first parts of messages with identical bits and pieces of text larded in the same order in the cryptograms begin to show. Painvin feels the oddity could most likely have resulted from transposed beginnings according to the same key; the identical tops of the columns of the transposition tableau. Painvin sections the cryptograms by timeframe:
He does this with 20 blocks to reconstruct the transposition key. Using the principle - long columns to the left, he finds segments 3,6,14, 18 to left. Balance clustered to right. Using other messages with common endings (repeated) He segments the columns to the left. Correctly? No. He uses 18 additional intercepts to juxtaposition 60 letters AA's, AD's, etc. Using frequency count, he finds a monoalphabetic substitution. He finds column 5-8 and 8-5 are inverted.
Painvin sets up a skeleton checkerboard - he assumes correctly the order to be side-top:
| A | D | F | G | X | |
| A | |||||
| D | e | ||||
| F | |||||
| G | |||||
| X |
Since the message was 20 letters, the order might be side-top, repeated, meaning side coordinates would fall on 1st, 3rd, 5th.. positions during encipherment, so he separates them by frequency characteristics. In 48 hours of incredible labor, Painvin pairs the correct letters and builds the checkerboard, solving the toughest field cipher the world had yet seen. A cipher that defends itself by fractionation - the breaking up of PT letters equivalents into pieces, with the consequent dissipation of its ordinary characteristics. The transposition further scatters these characteristics in a particularly effective fashion, while dulling the clues that normally help to reconstruct a transposition.
Solve these:
Ger-3. Kalenderblatt August. K2 (Sonne) BRASSPOUNDERQV FHOHIC ICMPC KQM IXWWM QW KML WFMPM KMI *IQLQHI, KMI *PHWKICMLWI, KFPML KQM "*PHWKIC- FOMI," KQM AMKML VMWIJP WXJP CQMLM VXMOMW.Ger-4. Ungerechtes Schicksal. Eng. K4 GEMINATOR
IRFJA DRGAI RAMRT VFAKF DLUFS UXABR ADSEQ DBHMR XBAIC KVELR JAVKV AFDJI HMBHP IEQII HMQEL JEIIA QGAUB SSAVJ AVIAQ GATVC KAIIC VJBAI AQGAD KVELA D. hints: (zum zw-; zimm-)Fre-1. French digraphic. Christmas Greeting. MON NOM
DBAAB AADBB BBBAB CABAA BBCDC ACCAA BABAC AABBD ACBAA AAACA CABAC BCCCB BAAAB IJGFG GKJGJ FFGJH JGFIK JFGFH GGFKG FGHKG FFGJJ GGJIK GJFJG JGFJH FGIIG KIKJF.hints: (noel, plus). Look out for disruption area in cipher square.
© 2004 Qsr Nrwn