Klaus Schmeh’s cryptography blog shows a new cryptogram from the library in Lawrence, Kansas, which is really weird to me because I live in Lawrence, Kansas, and go to the library there every weekend. Strange odds. Lawrence is a college town, home of Kansas University and a big college basketball town. There are about 100,000 people here and it is about 30 miles from Kansas City. It is a very peaceful and idyllic community and the city and county governments do a really good job of keeping the community safe, clean and with many modern amenities, including the public library. No better place to live, in my opinion, and I am very happy here.
Doranchak posted the cryptogram, which was written in the pages of a library book here.
http://scienceblogs.de/klausis-krypto-k … s-library/
Although this is a separate discussion apart from the website of the post, I like this forum because you can post pictures and stuff. Here are the three parts of the cryptogram written on three different pages of the book:
UPDATED TRANSCRIPTION 5/12/2019
Part 1
YBMFPTYSYSNAOSBYAACPTM YSNEMTMOIYJGTEBYNIAJ APSNDETMYLGAIdISYMI HFBYTFAAWOTYCMCHNEFTY IHSAWOYYHNEAWOWIWLTDI HAJBSASTFYTMMFdOETTD ITWYGMTIPDTTOFYANRSITL OACYANWYAJAHCTDSIHND ONIPEICNCOACOHMOOOT YPMDISWtOKIEYWOP,R,A AOTWFFFYOTYOMIHLYD WICSTOTIhhNCFBYHNLAO WALYCWOWTDSOLIOMWWW YNTLTYMYFYLNAWNMWSSI NDIDCWGOOBOYAYOFW GYKODLIAALACNSuFMSIAt OJBaHTYIANaWJSWDET OMYHABEODIASMGWIhdt TOTYHNdTIAPASYOAYdTS WBAAF,YFNSmTMOTOHYA NSYJDWMABTdEKMMSA IHO,M,LTPYANWADNWTSWIS AIHNGTSTBFANTdWL.
Part 2
IAAYYAARYNTTMIFLYTIA BYWYEWGOTAYNTTMWITA COSSOCAYSLIYATMTTOTT NAAMDOTWYNSOETBRNBAH TMAYSIDKINADRHBOFTW TTFBWSYJSBYKIWBMATO GTDALAYANIATADWCBYD DLOAHWIAFHWSTAYSTF OHSELTODTNOAMaHMA HWCUALFMTFYM.YNWTGTO G.AHTCHASOD.ATRISWNB. WISIAAOHB,BYCSTSIPYAA FITIYWSTASIABBaagi WSYCIATOYSIAWyd,BHS AIIMYMMFUIOIYRYCH CLEFAOWNHOTSADIHIW YCTTACBIWLFYTTiiM,L PIDOE,GSSIiMaTCIDLS DMSFHATHTLTIFTNWIA TOYLMJOMATLOIMFJMW YTOOMdYLDBLPCEAT
I haven’t pursued this aggressively. I had to work on the house this weekend. Except that I did look at frequency. Left, the first part of the message frequencies only 445 ciphertext. Right, a 32,640 chunk of my short stories plaintext. Some of the frequencies fairly match up, left to right, and there is a group that matches up but not aligned. I thought about a keyword simple substitution, but that would cause the group on the left to go down on the bar chart, not up.
Here, the blue is the 32,640 chunk, and the red is the 445 message, sorted by frequency high to low. Is is a pretty close match.
I looked into this cipher before reading the discussion at Klaus and dismissed the idea of it being a first-letter cipher because of the higher chi-square (lower is better). Which seems to contradict doranchak’s results at https://docs.google.com/spreadsheets/d/ … sp=sharing Though I am using the same frequency tables from Wikipedia.
Chi-square versus English: 537.14
Chi-square versus first-letter English: 810.47
A match would usually fall between 50 and 100 so this is a large difference! I thought it could be that first-person language has more "You" and "I" but after reviewing a first-letter 408 plain text it seems unlikely?
ilkpbiismfiimftkwgit fbmitmdaoatksgmtmtei iebtgyrowagtbpoiitwi diwbripaatihkwbmsiwn gymnbywttsdosmcosfma
Chi-square versus English: 115.24
Chi-square versus first-letter English: 78.39
The average per letter frequency difference when compared to expected distribution of first letters is 1.30%, compared to 1.95% when compared to expected distribution of all letters. I wonder if the difference is actually not that significant especially in light of the big difference in chi^2 test.
Double letter frequency is above average, imo, for English language. Also, following double letters, no -ING structure can be found as repeating. Last line of first part is almost equal to the single line (second part) of the cipher.
‘OTY’ and ‘TDI’ is the maximum repeating trigrams, both occurring only twice e.g. in the first part of the cipher (~450 symbols). Not enough to cover "AND" or "THE" for cleartext.
Surely not a simple substitution cipher.
QT
*ZODIACHRONOLOGY*
I wondered if the short one from the second page might be a transposition key of some sort. There are cryptography books in the junior section of the library. I have read them. Maybe a junior cryptographer with one of those books just having fun. Applying the short one to keyed columnar transposition might be worth a try.
EDIT: And then simple substitution.
I wondered if the short one from the second page might be a transposition key of some sort. There are cryptography books in the junior section of the library. I have read them. Maybe a junior cryptographer with one of those books just having fun. Applying the short one to keyed columnar transposition might be worth a try.
EDIT: And then simple substitution.
The word THE should occur approximately 4.8 times…without THEre, THEy, THEm, THEse etc…max repeating trigrams in the text (without the middle one, which is nearly double) is 2 (considering no double letters in such, like the word THE has no double letter either). Therefore, the text had a maximum of two times the word THE (instead of 4-5).
It’s still possible that it is a simple substitution, but then it’s a statistical outlier. Also, there are way too many double letters, imo (‘lyrical’ text/author?). I also wonder why there are small letters (e.g. d, a) although this wouldn’t be necessary for a simple substitution (figures?).
If the following section
YNWTGTOG.AHTCHASOD.ATRISWNB.WISIAAOHB
(lengths: 8, 9, 8, 9)
is actually meant to represent words: They do not match to the ENGLISH language..eg for the first one OPTIMISM or OBLIGING doesn’t match the last one (WISIAAOHB: besetting, caballero, colonnade, decennial, gonorrhea, necessary, necessity, perennial, rebellion, recession) as none of those words actually start with either an L or a T (represented by the W symbol)..so IF it was a substitution cipher: Different language?
A supposed to be complete list for the first two words (not necessarily English):
achtstes lotkolben
chalklik rolvormig
coprirsi berkebast
dyslalia rolvormig
gonimium heidehaus
gyakukou sekresjon
hyoujunj meulemans
hyoujunj sauraseni
hypereur adeodatus
hypoioni egorgeant
imposons egorgeant
laminion reigerbos
lycodoid egorgeais
lycodoid egorgeait
lycodoid reoperais
lycodoid reoperait
ouvrirai berkeblad
patinion reigerbos
shelflif rolvormig
shokikei alkylated
shutitei lotkolben
sourirai berkeblad
stalklik rolvormig
stormram berkeblad
tsumamia lemfeldig
tsumamia lempelziv
wehrprop carmacion
wehrprop surfusion
whelklik polyopsia
whelklik rolvormig
zaliging reiterons
For the last two words:
ladeprog repelling
lagorche corollike
lagorche corolline
latebris rebellons
lysebrun rebellion
satinflo finissero
schamrot ramassent
subentro tenessimo
zomerdag derezzing
But do those not necessarily match to each other..only such combination would be cleartext, but is there actually – none. At least not with the main languages of our world (it still might be some Suaheli dialect..).
Thus, even considering words such as GONIMIUM or HEIDEHAUS, this specific section would not fit to a simple substitution, imo.
However: There still is the possibility of a substitution, e.g if the author of the cipher had used e.g. only every second letter. Or it was rearranged (transposition) in some kind of way before writing it down as sort of fluent text. That also could explain the higher than average amount of double letters..
This is all so much Babylonian language confusion..
QT
*ZODIACHRONOLOGY*
I asked some ACA folks to try out their identification tools on it:
http://cryp.ackgame.com/2019/04/23/unkn … comment-79
The Rat’s response:
I don’t think it’s a transposition type. The Normor score is too high. The W and Y are too frequent. The index of coincidence is consistent with normal English and a Patristocrat came out on top in my Analyzer with a long stretch in the middle of this. It’s not clear what to do with the lower case letters. My program isn’t designed to analyze anything like that since no ACA cipher mixes them. There’s no periodicity so that leaves out a lot of types. If I had to guess I’d go for either a hoax or a Null.
The null or concealment cipher idea is intriguing. I wonder if there is some rule that will exclude the extraneous text and leave a real plaintext message.
Another response:
I tried a neural net collection that uses all of my experimental stats and it got:
Top cipher types were:Amsco with 14 votes
Patristocrat with 14 votes
CheckerBoard with 2 votesTotal votes: 30
But, as RAT says, it has too many Y’s (74) and W’s (52) and too few E’s (17), to be convincing as a transposition cipher. And it doesn’t have enough repeated digraphs to be convincing as a simple substitution cipher.
The mnemonic theory seems as good as any.
What is Normor score?
A Patristocrat can be ruled out:
Aristocrat and Patristocrat are ACA terms for simple substitution ciphers in which no letter is substituted for itself. Aristocrats have word divisions, Patristocrats do not.
As I mentioned earlier to doranchak there are some interesting "repeats" of the same form "FYAN", "CYAN", "FYCN", "TYIAN", "PYAN", "TYHN", etc.
As expected with the low bigram count, the cipher tested negative with AZdecrypt versus many different languages and English with spaces.
I ran my transcription of the first part through my 5-gram fragments measurement and here are the results:
Period 1: 7444, 7444 (3.85, 3.85) <--- Period 2: 7170, 7234 (2.47, 2.79) Period 3: 6684, 6504 (0.03, -0.87) Period 4: 6768, 6626 (0.45, -0.26) Period 5: 6646, 6682 (-0.16, 0.02) Period 6: 6520, 6704 (-0.79, 0.13) Period 7: 6954, 6912 (1.38, 1.17) Period 8: 6592, 6368 (-0.43, -1.56) Period 9: 6606, 6772 (-0.36, 0.47) Period 10: 6810, 6788 (0.66, 0.55) Period 11: 6452, 6848 (-1.13, 0.85) Period 12: 6756, 6682 (0.39, 0.02) Period 13: 6598, 6426 (-0.40, -1.26) Period 14: 6874, 6830 (0.98, 0.76) Period 15: 6534, 6790 (-0.72, 0.56) Period 16: 6774, 6648 (0.48, -0.15) Period 17: 6490, 6458 (-0.94, -1.10) Period 18: 6874, 6684 (0.98, 0.03) Period 19: 6490, 6942 (-0.94, 1.32) Period 20: 6450, 6608 (-1.14, -0.35)
Peaks at period 1 and 2. A score of 7444 happens only once in about 2750 shuffles so these peaks appear to be real. I agree that transposition can be ruled out.
The null or concealment cipher idea is intriguing. I wonder if there is some rule that will exclude the extraneous text and leave a real plaintext message.
Indeed.
I cannot get the first-letter hypothesis out of my head. Could it still be that but in another language? Will have to check chi-square stats versus different languages, feel free to do this work if you can, I will not be able to check it within 16 hours or so.
Someone probably needs to rule out the obvious: Is the source of the mnemonic the very book it is written in?
Maybe someone was practicing memorization of long passages and used the mnemonic as an aid.
What is Normor score?
It is a test devised by "THE RAT":
http://www.ackgame.com/Normor%20Revisited.pdf
It seems similar to calculating the distance between the observed and expected distributions of letters.
As I mentioned earlier to doranchak there are some interesting "repeats" of the same form "FYAN", "CYAN", "FYCN", "TYIAN", "PYAN", "TYHN", etc.
That is a very interesting observation. Basically, there seems to be a lot of pairs of sequences that have short edit distances. For example, the edit distance between CYAN and PYAN is one because only one correction is needed to make CYAN into PYAN.
I looked for all such patterns, of length at least 4, having an edit distance of only one. Results:
https://docs.google.com/spreadsheets/d/ … sp=sharing
There are some interesting longer examples, like:
YNTTMI and YNTTMWI
ARYNTTM and AYNTTM
WSTAYS and WSTAS
etc
Total list was about 500 patterns. But in a random shuffle, there were 400 similar patterns. So I’m not sure how significant this is. However, in the shuffle, there were fewer of the longer patterns.
It may be interesting to cluster all of them together rather than considering only pairs. Perhaps the clusters are more significant than the quantity of individual pairs.
Here are longer sequences that match with edit distance of 2:
89 236 YHNEAWOW YHNLAOW 506 610 TOTTNAAM TODTNOAM 459 481 YNTTMIF YNTTMWIT 459 481 YNTTMI YNTTMWIT 459 480 YNTTMI AYNTTMWI 458 480 RYNTTMI AYNTTMWI 457 480 ARYNTTMI AYNTTMWI 457 480 ARYNTTMI AYNTTMW 457 480 ARYNTTMI AYNTTM 456 480 AARYNTTM AYNTTM 456 479 AARYNTTM TAYNTTM 44 332 SNDETMY SWDETOMY 134 390 TOFYANRS TOHYANS 90 237 HNEAWOW HNLAOW 89 236 YHNEAWO YHNLAO 89 236 YHNEAW YHNLAOW 706 810 IATOYSI IATOYLM 706 810 IATOYSI IATOYL 706 810 IATOYS IATOYLM 706 809 IATOYS WIATOYL 705 810 CIATOYS IATOYL 705 809 CIATOYS WIATOYL 705 809 CIATOYS WIATOY 705 809 CIATOY WIATOYL 705 808 CIATOY NWIATOY 704 809 YCIATOY WIATOY 704 808 YCIATOY NWIATOY 70 212 OTYCMCH OTYOMIH 674 688 YCSTSI YWSTASI 597 689 WSTAYST WSTASI 596 688 HWSTAYS YWSTAS 59 463 MIHFBYT MIFLYT 53 721 AIdISYM AIIMYM 507 611 OTTNAAM ODTNOAM 506 610 TOTTNAA TODTNOA 506 610 TOTTNA TODTNOA 460 482 NTTMIF NTTMWIT 46 334 DETMYL DETOMYH 459 481 YNTTMIF YNTTMWI 459 481 YNTTMIF YNTTMW 459 480 YNTTMI AYNTTMW 458 481 RYNTTMI YNTTMWI 458 481 RYNTTMI YNTTMW 458 480 RYNTTMI AYNTTMW 458 480 RYNTTMI AYNTTM 458 480 RYNTTM AYNTTMW 458 479 RYNTTM TAYNTTM 457 480 ARYNTTM AYNTTMW 457 479 ARYNTTM TAYNTTM 456 479 AARYNTT TAYNTT 45 333 NDETMY WDETOMY 44 332 SNDETM SWDETOM 4 19 PTYSYSN PTMYSN 37 574 YNIAJA YANIATA 315 665 SIAtOJB SIAAOHB 182 212 OTYPMDI OTYOMI 148 421 YANWYAJ YANWAD 148 393 YANWYAJ YANSYJ 147 420 CYANWYA PYANWA 135 391 OFYANRS OHYANS 134 390 TOFYANR TOHYANS 134 390 TOFYANR TOHYAN 134 390 TOFYAN TOHYANS 134 389 TOFYAN OTOHYAN 133 390 TTOFYAN TOHYAN 133 389 TTOFYAN OTOHYAN 459 481 YNTTMI YNTTMWI 457 480 ARYNTTM AYNTTM 89 236 YHNEAW YHNLAO 88 235 YYHNEA BYHNLA 87 357 OYYHNE OTYHNd 86 356 WOYYHN TOTYHN 71 213 TYCMCH TYOMIH 707 811 ATOYSI ATOYLM 706 809 IATOYS WIATOY 705 810 CIATOY IATOYL 704 808 YCIATO NWIATO 70 212 OTYCMC OTYOMI 70 182 OTYCMC OTYPMD 69 211 WOTYCM YOTYOM 69 181 WOTYCM OOTYPM 68 85 AWOTYC AWOYYH 67 84 AAWOTY SAWOYY 636 808 NWTGTO NWIATO 621 644 AHWCUA AHTCHA 62 465 FBYTFA FLYTIA 609 803 LTODTN LTIFTN 6 268 YSYSNA YFYLNA 597 689 WSTAYS WSTASI 597 688 WSTAYS YWSTAS 596 688 HWSTAY YWSTAS 595 687 FHWSTA IYWSTA 509 613 TNAAMD TNOAMa 508 612 TTNAAM DTNOAM 507 611 OTTNAA ODTNOA 506 610 TOTTNA TODTNO 506 566 TOTTNA TOGTDA 505 609 TTOTTN LTODTN 504 564 MTTOTT MATOGT 479 599 TAYNTT TAYSTF 478 598 OTAYNT STAYST 478 515 OTAYNT OTWYNS 477 514 GOTAYN DOTWYN 460 482 NTTMIF NTTMWI 46 334 DETMYL DETOMY 459 482 YNTTMI NTTMWI 459 480 YNTTMI AYNTTM 458 481 RYNTTM YNTTMW 457 480 ARYNTT AYNTTM 457 479 ARYNTT TAYNTT 45 334 NDETMY DETOMY 448 663 WL.IAA WISIAA 44 332 SNDETM SWDETO 43 331 PSNDET JSWDET 429 484 TSWISA TMWITA 427 482 NWTSWI NTTMWI 421 574 YANWAD YANIAT 420 573 PYANWA AYANIA 410 433 SAIHO, SAIHNG 409 432 MSAIHO ISAIHN 408 431 MMSAIH WISAIH 387 503 TMOTOH TMTTOT 386 502 mTMOTO ATMTTO 38 576 NIAJAP NIATAD 374 439 TSWBAA TSTBFA 374 429 TSWBAA TSWISA 37 575 YNIAJA ANIATA 356 390 TOTYHN TOHYAN 336 356 TOMYHA TOTYHN 335 355 ETOMYH tTOTYH 316 666 IAtOJB IAAOHB 315 665 SIAtOJ SIAAOH 314 664 MSIAtO ISIAAO 24 384 NEMTMO NSmTMO 23 44 SNEMTM SNDETM 216 463 MIHLYD MIFLYT 215 462 OMIHLY TMIFLY 182 212 OTYPMD OTYOMI 181 211 OOTYPM YOTYOM 179 388 MOOOTY MOTOHY 175 489 ACOHMO ACOSSO 174 615 OACOHM OAMaHM 15 681 YAACPT YAAFIT 148 421 YANWYA YANWAD 148 420 YANWYA PYANWA 148 393 YANWYA YANSYJ 147 420 CYANWY PYANWA 147 392 CYANWY HYANSY 146 419 ACYANW TPYANW 135 391 OFYANR OHYANS 134 356 TOFYAN TOTYHN 133 389 TTOFYA OTOHYA 132 334 DTTOFY DETOMY 122 514 DITWYG DOTWYN 121 513 TDITWY MDOTWY 706 810 IATOYS IATOYL 705 809 CIATOY WIATOY 459 481 YNTTMI YNTTMW 458 480 RYNTTM AYNTTM 134 390 TOFYAN TOHYAN
I looked for periodicity in bigrams, coincidence count unigrams, and a difference between bigram repeats starting in odd positions versus even positions. Never thought that it was a vigenere or digraph cipher because of the frequencies though. And no results for any. I poured the first page into a grid just in case the second page was a key for keyed columnar transposition, but not solve.
The "key":
A I H N G T S T B F A N D T N L
And the message poured into the key:
Y Y P O R I T Y I Y M N S N C T
M W E L S A O A S T O D A W N O
B O I I I P M N Y M I I W A S T
F P C O T A Y S M M Y D O D U I
P R N M L S H Y I F J C Y N F H
T A C W O Y A J H D G W Y W M H
Y A O W A O B D F O T G H T S N
S O A W C A E W B E E O N S I C
Y T C Y Y Y O M Y T B O E W A F
S W O N A D D A T T Y B A I T B
N F H T N T I B F D N O W S O Y
A F M L W S A T A I I Y O A J H
D F O T Y W S D A T A A W I B N
S Y O Y A B M E W W J Y I H A L
B O O M J A G K O Y A O W N H A
Y T T Y A A W M T G P F L G T O
A Y Y F H F I M Y M S W T T Y W
A O P Y C Y H S C T N G D S I A
C M M L T F D A M I D Y I T A L
P I D N D N T I C P E K H B N Y
T H I A S S T H H D T O A F A C
M L S W I M O O N T M D J A W W
Y Y W N H T T M E T Y L B N J O
S D T M N M Y L F O L I S T S W
N W O W D O H T T F G A A D W T
E I K S O T N P Y Y A A S W D D
M C I S N O D Y I A I L T L E S
T S E I I H T A H N D A F
Someone probably needs to rule out the obvious: Is the source of the mnemonic the very book it is written in?
Maybe someone was practicing memorization of long passages and used the mnemonic as an aid.
I found a pdf of the aforementioned book, and wrote a quick script looking for word sequences that would match mnemonic sequences in the cipher, and couldn’t find anything. It wasn’t exhaustive but nothing even matched the initial sequence "YBMF", so that can probably be ruled out.
