What mistakes did Scribe 2 never make?

by ⋅ 47 Comments

In her recent (2020) Manuscript Studies paper “How Many Glyphs and How Many Scribes? Digital Paleography and the Voynich Manuscript“, Lisa Fagin Davis builds up to the conclusion (p.179) “The fact that all of these collaborative methods involve Scribe 2 may suggest that she or he was in charge of the project in one way or another.“

If this is the case, then I think the implicit question it suggests we ask is: what mistakes did Scribe 2 never make? That is, if Scribe 2 ‘knew what she or he was doing’ with Voynichese more than Scribes 1 and 3-5, we might sensibly expect Scribe 2 to make fewer scribal errors than the others. So, might we be able to use this prediction to tell good Voynichese (well, Currier B-ese, anyway) apart from miscopied Voynichese, hmmm?

The list of places where we can find Scribe 2 is as follows:

  • All the Herbal B pages (apart from the f41-f48 bifolium, which was written by Scribe 5)
  • The entire Q13 Balneo section
  • One side of the nine rosette foldout (Scribe 4 wrote the other side)
  • The first 12 lines of f115r (everything else in Q20 was written by Scribe 3)

One general problem with Voynichese is that – contrary to the wisdom of much of the Internet – it isn’t quite a game of two halves, i.e. a Currier A half and a Currier B half. Within those distinct variants, individual sections vary yet further: so, even though Q13 and Q20 are both ‘Currier B’, each one’s use of Currier B presents plenty of differences from the other. So if we are looking for differences, we have to be careful not to get caught up in the subtleties of how the (for want of a better word) style of Voynichese itself shifts between sections.

As a result, the two specific comparisons I think we should interested in here are the Herbal B pages (i.e. how does Scribe 5’s use of Voynichese differ from Scribe 2’s?) and the Voynichese on f115r (i.e. how does Scribe 3’s use of Voynichese differ from Scribe 3’s?). Let’s dive in and have a closer look…

Herbal B: Scribe 5 vs Scribe 2

The issue here is essentially comparing Scribe 5’s writing on f41 and f48 with Scribe 2’s writing on f26, f31, f33, f34, f39, f40, f43, f46, f50 and f55. Sadly, voynichese.com only offers a single filter of Currier A vs Currier B pages, which makes it not quite as useful as it might be (i.e. we’d like to do tests on [Herbal B + Scribe 2] vs [Herbal B + Scribe 5]). Maybe someone will add an LFD Scribe filter at some point in the future. 😉

But there is yet another dimension of difficulty to throw into the mix: transcription ambiguities. Because transcribers have quite a torrid time distinguishing characters (e.g. “a” vs “o” vs “y”, “cc” vs “ch”, “sh” vs “se”, and please don’t get me started on half-spaces vs spaces, *sigh*), we have to be careful we don’t mistake a transcriber’s whim for a scribal tell.

So the way I started was by grabbing the Takahashi transcriptions for f41r (Scribe 5) and f26r (Scribe 2), and comparing them really closely to high resolution images (on Jason Davies’ Voyage the Voynich website). My plan was to try to get a feeling for whether there was any visual evidence that indicated Scribe 2 was an author (i.e. who understood the internal construction of Voynichese) and the other just a dumb scribe (i.e. who was just copying what they saw).

However, I quickly found a fair few examples of what seemed (to my eyes) to be basic Voynichese scribal errors by Scribe 2.

  • f26r line 2: second word looks like it should be “daiin”, but the first glyph is somewhat malformed
  • f26r line 2: third word “adeeody” looks like a scribal slip for “odeeody”
  • f26r line 2: free-standing word “lr” looks like a scribal slip for “ar”
  • f26r line 3: word-terminal “-oy” looks like a scribal slip for “-dy” or possibly “-ey” (particularly in Currier B, though “qoy” is probably OK)
  • f26r line 3: shapchedyfchy looks like a scribal slip for shopchedyfchy
  • f26r line 3: penultimate word “saiin” looks like a scribal slip for “daiin”

By way of contrast, Scribe 5’s writing – though typically a little harder to transcribe – was generally quite clear, without any obvious scribal errors. So I would say that comparing these two pages (while only a relatively small sample) offers no obvious support to the notion that Scribe 2 might have had a more authorial understanding of Voynichese. On the contrary, it seems more likely to me from this that Scribe 2 was, well, just a scribe.

f115r: Scribe 3 vs Scribe 2

The first twelve lines of f115r (that Lisa Fagin Davies attributes to Scribe 2) present what look to me like yet more scribal errors by Scribe 2. For example:

  • f115r line 1: “oechedy” (a hapax legomenon) looks like a scribal slip for “orchedy”
  • f115r line 1: “oroiir” looks like a scribal slip for “oraiir”
  • f115r line 3: the penultimate word “daar” (another hapax legomenon) looks like a scribal slip (possibly for “-dy ar-“)
  • f115r line 3: the final word “oraro” looks like a scribal slip for “orary”.
  • f115r line 5: the final word “ro” looks like a scribal slip for “ry”
  • f115r line 12: the final word “choloro” should probably be “cholory”

(Incidentally, I should note that 12 out of all 13 instances of the word “ry” appear right at the end of a line [the other one appears right at the start of a line]. There is no shortage of patterns in Voynichese on all sorts of levels!)

Again, the remainder of f115r (attributed to Scribe 3) seems basically OK, so Scribe 2 again seems to be copying in the letters quite a lot worse than Scribe 3.

So… Scribe 2 was not the Voynich’s author, right?

From all this, it’s looking to me as though we can infer that Scribe 2 was not the author: or, more precisely, that Scribe 2’s errors seem consistent with the idea that Scribe 2 had no authorial level of understanding of the internal structure of Voynichese. Which, of course, would seem to be the opposite of what Lisa Fagin Davis’ paper suggested (if you read its conclusions in the strongest way possible).

However, I think this does imply something quite deep about the reliability of different sections of the Voynich, which is that some would seem to be less tainted by scribal errors than others. Though based on what is only a small sample, I suspect that Scribe 2 is a more unreliable Voynich scribe than both Scribe 3 and Scribe 5.

For a long time, I’ve been telling Voynich researchers that they should avoid treating the whole of the Voynichese corpus as if it were a single coherent text (because it isn’t): and that they should instead run their statistical analyses on individual sections, such as Q13 and Q20. However, because Scribe 2 wrote the entirety of Q13, I’m now revising that opinion: my particular concern is that Scribe 2’s copying errors (and I’ve only highlighted the errors I can see, there could easily be many others I can’t see) might well enough to disrupt any statistical studies.

Hence my recommendation going forward is that researchers should focus their decryption attempts on Q20, specifically excluding the top twelve lines of f115r (written by Scribe 2).

Why did Scribe 2 write the top part of f115r?

Might there have been a good reason why Scribe 2 wrote the top few lines of f115r?

Possibly. I’ve blogged a number of times about Q20 (which contains far too many bifolia to be a single quire), and how I think it may originally have been constructed as two separate gatherings Q20A and Q20B. The fancy gallows at the top of f105r looks a pretty good bet to have been the start of Q20A, and the current back page (f116v) similarly looks a good bet to have been the end of Q20B. I also wondered whether f104, f105, f107 and f108 may all have been cut from the same piece of vellum (an hypothesis which could at least be tested using DNA now).

(As an aside, I suspect that the seven dots on f105r imply that this marked the start of “Liber VII”. Just so you know.)

All of which would seem to point away from the (long-standing) suspicion that Q20 was written as a single monolithic slab, and instead towards the suggestion that Q20 / Q20A / Q20B might well have included separate sections. Might the first few lines of f115r have been written as the start of a section? Or might it even have been the end of a section, running on from a previous page (say, in Q8)? These are all no more than suggestions at this stage, make of them what you will. Possibly a nice risotto.

Q20 bifolio content notes

Finally: as Rene Zandbergen pointed out in 2016, the paragraph stars for the majority of f111r look to be fake. Yet the same seems true for the top half of f111v, as well as the bottom half of f108v, and the middle third of f115r; and there’s a paragraph star apparently missing from the middle of both f106v and f113r (though the latter of these two might possibly have just slid down the page). So we have to be extraordinarily careful when we try to draw inferences about the original section structure of Q20 based on the paragraph stars.

Here’s a brief summary for anyone trying to figure out the original nesting arrangement:

  • f103: both recto & verso have no ‘x’ character (Tim Tattrie)
  • – f104: recto has non-repeating star pattern (Elmar Vogt)
  • – – f105: recto has fancy gallows at top, possibly start of Liber VII?
  • – – – f106: verso has single paragraph star missing
  • – – – – f107:
  • – – – – – f108: recto has non-repeating star pattern (Elmar Vogt); verso has fake paragraph stars on bottom half
  • – – – – – f111: recto & verso both have many fake paragraph stars (Rene Zandbergen)
  • – – – – f112: both sides have a gap by the outside edge (possibly a copy of a stitched vellum tear, cf Curse 2006)
  • – – – f113: recto has single paragraph star missing
  • – – f114:
  • – f115: recto has Scribe 2 writing at top (Lisa Fagin Davis) & fake paragraph stars in middle third
  • f116: recto has no ‘x’ character, verso has michitonese + pen trial doodles

Somerton Man: It’s Edge of the Seat Time…

by ⋅ 19 Comments

Like a constipated true crime podcaster, I’m currently perched on the edge of my seat waiting for something solid to emerge. Now that SAPOL’s forensic finest (surely) have the Somerton Man’s DNA in their sweaty hands, what will it be able to tell us?

One interesting thing about DNA searching is that even if you get basically zero hits, the DNA itself can often still tell you a great deal about a person, such as:

  • what part of the world they (probably) come from
  • their haplogroup (& will that be the same as the haplogroup Derek Abbott’s group retrieved from the hair root?)
  • their genetic predisposition to rare illnesses (e.g. uncombable hair syndrome, etc).

If the part of the world the Somerton Man’s DNA comes from is basically a small region in Ireland, it would seem to be a fairly strong indication that Kean[e] is likely to be his surname. (But with Catholic families being DNA genealogists’ best friends, you’d also expect 20+ decent hits to light up the GEDmatch globe like a Christmas tree.)

Yet if his DNA is solidly Eastern European (and with hardly any matches), you’d expect a quite different person – perhaps something like the mysterious Balutz from the baccarat school I found so hard to track down.

Though it would be nice if the DNA showed he was Charles Mikkelsen (who I think was probably also the “Carl Thompsen” remembered by Keith Mangonoson), I’m not holding out a lot of hope for that.

It also seems likely to me that any link to the Abbotts / Egans would have been trumpeted to the world’s media by now: but given the lack of trumpetry my ears are picking up, this is most probably not to be here.

All in all, it’s perhaps surprising that the list of possible Somerton Man candidates we’ve all managed to accumulate is so short: a list dominated, it has to be said, by implausible Soviet spies, defectors and perhaps even spring-heeled Ballet Russe dancers. (Spare me, O Lord, from having to read any more espionage-related posts.)

So I wonder what the next card to be played in this interminable squeeze will be?

Henry Debosnys: Crowdfunding DNA Extraction…?

by ⋅ 8 Comments

For the last couple of weeks, I’ve been trying to work out if there was a way I could set up a funding page for a genealogist-driven collaborative effort to identify Henry Debosnys from his DNA.

Obviously, the funding would need to cover the two physical stages:

  • Extracting his DNA – the nice people at the Adirondack History Museum have both his skull and the hangman’s noose that killed him (the skin on his neck was abraded for four inches, according to a contemporary newspaper report), so this looks to be achievable; and
  • Scanning the extracted DNA and uploading it into some historical database for familial matches (the same as virtually every other programme on TV at the moment).

It turns out that extracting his DNA is the (relatively) easy part. The second part is actually much trickier, because it seems that pretty much all the consumer services out there (e.g. AncestryDNA, even Nebula Genomics) have highly optimised their low-cost DNA acquisition pipelines for either cheek-swab / saliva samples or blood samples.

Moreover, the helpful support people at AncestryDNA also told me that it would explicitly go against the company’s policy to take DNA samples from someone who had died. So there would seem to be a series of mountains for historical researchers to climb there.

Note that there are a (very small number of) commercial whole genome scanning services out there that don’t rely on cheek swabs or blood: but my understanding is that these tend to be quite expensive. So as of right now, it would seem that we’re kind of stuck between the two: crowdfunding a cheek swab test would be do-able (probably south of 1000USD, all in), but I’m guessing that this would rise to about 2500USD with a bespoke special WGS from extracted DNA.

Having recently spent time going through French archives via filae.com, I had thought that trying to track Debosnys’ genetic footprints would be a great project to crowdfund and take on, but I’ve been left somewhat bemused. So if anyone reading can suggest a better route forward for scanning extracted DNA and then GEDmatching it, please leave a comment below, I’m all ears!

Finally, I should mention an alternative route. Australian DNA genealogy company totheletter DNA have been offering a (really rather incredible) service where you send them old letters / stamps and they then extract DNA from the saliva used to stick the dried adhesive down. However, it turns out that they’ve been having some problems with the quality of the DNA extracted in this way, so they temporarily halted the service last year, but hope to bring it back online later this year (2021).

Might Paul Olum have been the person behind Richard Feynman’s Ciphers?

by ⋅ 5 Comments

A fascinating email from long-time Cipher Mysteries commenter Paul Relkin has alerted me to a pair of ciphers by mathematician Paul Olum (1918-2001), who knew Feynman at Princeton, and then worked with him at Los Alamos. Sure, you can read Olum’s Wikipedia page, but perhaps the best person to describe him is Richard Feynman himself (in “Surely You’re Joking, Mr Feynman”):

They were all giving me problems and I was feeling great, when Paul Olum walked by in the hall. Paul had worked with me for a while at Princeton before coming out to Los Alamos, and he was always cleverer than I was. For instance, one day I was absent­mindedly playing with one of those measuring tapes that snap back into your hand when you push a button. The tape would always slap over and hit my hand, and it hurt a little bit. “Geez!” I exclaimed. “What a dope I am. I keep playing with this thing, and it hurts me every time.”

He said, “You don’t hold it right,” and took the damn thing, pulled out the tape, pushed the button, and it came right back. No hurt.

“Wow! How do you do that?” I exclaimed.

“Figure it out!”

For the next two weeks I’m walking all around Princeton, snapping this tape back until my hand is absolutely raw. Finally I can’t take it any longer. “Paul! I give up! How the hell do you hold it so it doesn’t hurt?”

“Who says it doesn’t hurt? It hurts me too!”

I felt so stupid. He had gotten me to go around and hurt my hand for two weeks!

So Paul is walking past the lunch place and these guys are all excited. “Hey, Paul!” they call out. “Feynman’s terrific! We give him a problem that can be stated in ten seconds, and in a minute he gets the answer to 10 percent. Why don’t you give him one?”

Without hardly stopping, he says, “The tangent of 10 to the 100th.” I was sunk: you have to divide by pi to 100 decimal places! It was hopeless.

One time I boasted, “I can do by other methods any integral anybody else needs contour integration to do.”

So Paul puts up this tremendous damn integral he had obtained by starting out with a complex function that he knew the answer to, taking out the real part of it and leaving only the complex part. He had unwrapped it so it was only possible by contour integration! He was always deflating me like that. He was a very smart fellow.

Olum I

But back to the recently unearthed pair of Paul Olum’s ciphertexts. The first one surfaced in 2018 at a Caltech exhibition to celebrate what would have been Feynman’s 100th birthday, and then bubbled through into a Reddit post in 2019. This (marked as “Olum I”) included Feynman’s notes on it (character frequency counts etc):

Feynman’s conclusion was that it was probably a “simple substitution [cipher] with some rearranging. The rearranging is not very complete“, and suggested some possible substitutions: “N -> T, G -> K, X -> N, F -> C, D ->G, (U -> W)?“. Though, as seems to have been the case fairly often, he didn’t quite get the better of Olum here.

If you want to have a tilt at Olum’s first windmill, the Reddit poster (“V sbyybjrq gur ehyrf”) transcribed its 744 characters as follows:

VEWLJ NBBEL QFWSX HBUSW AIBYQ AEQSI GHOVN BSNBV LNWXA BIQIU BGBIC YQFXC EVBWX WBSNG WVEVL HWDHB ILMHB LNSGB HSNXS XBHLQ CBOCS OBVWM XFNCW PAGGN EUWGA IBVWI FYWFG GQFEW MPQIX XWSEW VIHAB EBWJX OHAFQ LBBNI BHAIV JNSHC WXPCY UGGOB DWAXB HBWIN XWSNJ GWVAF OXBLM WAEBP BBWXC RBWBV LHIJA JINOW XDBIB QCGYW FXHCQ AIBCW NGCSC SHBNA VIEWD HIBLH EBVVY YSLRQ PQVCQ IWXQE DQBIW XWEAP BHVWS BSBWX VAVHB WFPUH BYWVN BYIOQ WAIFY QDXDB ICLBW YCNEA IBWIN BBWAA CIQIC VWIXQ VCBLH XIBVL AHMFO BXSIX OQBUE PCOVA WMOFV NCWAP GGNEE UWAIW XAWAE EWOLE WESHW FXHEG HCIVB HSWJO ILAWF NDDFQ WDHIL VHBBW AIQBI OUXWS BNIGW VXVQD BVAWI FGWXN VWEPU HYWDB HIMLH BLPNM WVHYP BYWBH AMFXX OSCVN BHCWV NYBIO QWAVI YBQXD LBNDV WCCGN AABXQ VWDBH EILHJ BLNVW VBHAF XOBCB MYWIN SBVOQ WLOHC GGWFB BNSYM DQUBX WSNGB WVWAI VGXHB OJWDB HIBVL HMWBI HIWJG NBFBH DQBIW IBOBJ OHUHV YLQMY WSNSI DFWDD VWEWV HYDLW VWGPW SSHAB ILBWY WJLHD XXSH

Olum II

Olum II (the second ciphertext Olum passed to Feynman) looks to be a pure transposition cipher, though anyone looking at its length (227, a prime number) and hoping for a simple modulo-227 transposition is going to be somewhat disappointed (I tried this to give my suffixity metric a bit of a workout). Note that the 25th character seems to be lower-case ‘f’, though whether or not that is relevant or useful is another matter entirely. Note also that, mathematically, there’s no point doing multiple consecutive modulo transpositions (e.g. modulo 6 followed by modulo 7), because an N-mod-A transposition followed by an N-mod-B transposition is the same as an N-mod-((A x B)mod N)) transposition:

The same Reddit poster (“V sbyybjrq gur ehyrf”) transcribed Olum II’s 227 characters as follows:

EEIOL CNTPA TIILM NIHGU TIGLf OOOHR BYSCD EYGSE EIEEL MERSB ITCBA ANEIT GDSDD OURDM SIOMH ESELE DNSRR NHNIN ATONW AEDSY ROWHE DRTRA SVAWH EODES ETVIF NIEHE TOIGI ELNII TONAR THTHL EULII TAISL SUNFC EAINI ELSLT LBPSN TMTIH SDSIH TREIE NDUET HHIOM EIIAS TVHPF YGSOR NEEII ET

To be honest, I would expect Paul Olum to have rolled out some kind of funky modulo maths trickery here, so my strong suspicion is that this is likely to be a test more of mathematical cunning than of cryptological brains.

Thoughts on this, Nick?

As Paul Relkin reminded me, I once crypto-profiled the author of the Feynman challenge ciphers as being most likely “interested in snickering into his beard about having pulled the wool over Richard Feynman’s sainted eyes“. And I think you’d have to admit that Paul Olum does seem to match that description well (errrm… apart from the fact he was clean-shaven). Yet proof is wondrously hard to achieve, so for now this remains no more than an interesting possibility.

As for Olum II: nowadays, I wonder whether the right place to start on this kind of complicated challenge transposition would be by searching for it in Project Gutenberg. By which I mean:

  • Get the A-Z frequency counts for the 227-letter challenge cipher
  • Go through all the Project Gutenberg files, converting them to A-Z (and no spaces)
  • For every 227-letter stretch in each file, compare the A-Z frequency counts against the cipher’s frequency counts
  • Display all the exact (and very close) matches you find (though right now I have no feel for how many that would be)

That is, I’m wondering whether you might be able to use Project Gutenberg to brute force an answer to a challenge transposition cipher without actually knowing how the transposition works.

This shouldn’t actually take long to calculate, and would parallelise very well. Anybody want to give this a go?

Somerton Man: “Three Main Theories”…

by ⋅ 30 Comments

A couple of days ago, I listened to a ten-minute online Somerton Man piece on Radio National Breakfast with Fran Kelly, basically because Fran had Gerry Feltus phoning in to give his tuppence worth. (Am I allowed to say that Gerry didn’t seem as Royal Sovereign H pencil-sharp as normal?)

As you’d expect, there wasn’t anything there of any great surprise or interest about the Somerton Man that you wouldn’t have picked up from even a cursory reading of Cipher Mysteries over the last few years. But the other person being interviewed – Fiona Ellis-Jones, who you may possibly remember as having been the host of the ABC’s five-part “The Somerton Man Mystery” podcast – did say one thing that I at least found interesting.

What she said (at 5:07) about the Somerton Man was this: that there were “three main theories: the love child theory; the fact that it could have been a black market racketeer; or perhaps a Russian spy“. Though this is basically rehashing her podcast tag line (“Was he a scorned lover? A black market racketeer? Or a spy?”), what struck me was that the whole black marketeer crim thing I’ve been pushing at for the last few years was suddenly in the top three.

Now, even though Fiona added that her own personal favourite theory was Derek Abbott’s whole love-child / spurned lover thang, it’s not exactly news that this has always seemed far too tidily romantic to me: all it’s lacking is a neat little bow on top, which is almost never how historical research actually works out. But the good news is that a DNA profile for the Somerton Man should make this the very first theory to be comprehensively disproved, all being well. :-p

As for the whole spy theory: apologies to John Ruffels etc, but if there’s an ounce of actual historical substance to that whole hopeful hoopla beyond “The Somerton Man is mysterious; spies are mysterious; therefore the dead guy must have been a spy“, I’ve yet to see it. Though it remains possible that the DNA match map will light up all across Russia, please excuse me if I seem less than utterly enchanted. Even vague familial DNA matches should be enough to rule out most of the exotic nonsense that some like to pass off as rock solid ‘fact’ (*choke* *cough* *cough*).

Moreover, if both those much-loved dominoes clatter to the floor, the question becomes: what other possibilities are we genuinely left with? Charles Mikkelsen (a favourite of Byron Deveson) remains ~vaguely~ possible, though it has to be said that Mikkelsen’s well-documented death at sea in 1940 does tend to spoil the party vibe there somewhat. Similarly, the 1953 death announcement for Horace Charles Reynolds that I (eventually) dug up doesn’t bode well for Somerton Man fans of a muttony disposition.

Might it be that the black marketeer theory might end up one of the very few realistic dominoes left standing before very long? Just thought I’d point that out… 😐

One Last Thing…

Something I noticed a few weeks ago was that even though I’ve posted 1490 blog posts on Cipher Mysteries since 2007-ish (originally as “Voynich News”), the times people have posted an actual link to anything I’ve posted are dwindlingly few. In fact, thanks to the magic of Google Search Console, I can tell you that Google knows of only 560 external links out there, many of which are repeated several times over. (“There may be many others but they haven’t been discarvard.“) Of those:

  • 113 are from labatorium.eus, all of which point to a page here on the Feynman challenge cipher (why?);
  • 89 are from voynich.ninja (mainly to Voynich-related pages);
  • 54 from blogspot.com blogs (most of which seem to be from numberworld.blogspot.com)
  • 35 from wordpress.com blogs (e.g. Koen’s herculeaf, Diane’s voynichrevisionist, and a handful of Rich’s proto57)
  • 20 each from voynichportal.com (thanks JKP) and voynichrevisionist.com (thanks Diane again)
  • 19 from reddit.com
  • 17 from scienceblogs.de (thanks Klaus)
  • 12 from zodiackillerciphers.com (thanks Dave O)

…while everything else is in single digits. How, then, has anybody ever found out about the black marketeer theory? Beats me.

Oh, and in case you’re interested, Cipher Mysteries’ pages include 7740 solid outbound links: which seems to imply I link roughly 20x more often outwards than everybody else combined links inwards. Perhaps it’s just me, but that statistic seems a bit sucky.

Just so you know how the Internet actually works.

Review: Benedek Lang’s “The Rohonc Code”…

by ⋅ 33 Comments

A copy of Benedek Lang’s nice-looking book “The Rohonc Codex: Tracing a Historical Riddle” landed on my doormat this week, courtesy of The Penn State University Press (its publisher). Its back cover blurb promises that it “surveys the fascinating theories associated with the Codex“, and that it finishes up by “pointing to a possible solution to the enigma“.

Though I was already a fan of Benedek (his (2008) “Unlocked Books” sits on the bookshelf just behind me), it was clear within a few pages of this new book that his (formerly densely academic) writing style has opened out in the intervening decade and a half. So anyone with an interest in the mysterious Rohonc Codex’s strange writing and pointy-chinned Biblical chappies will quickly find themselves drawn in to his accessible and readable account.

Benedek also partially presents the book as a sort of ‘survivor’s account’ of the wave of obsession with the Rohonc Codex that washed over him for a few years (which he was also fortunate enough to get grants to pursue). Perhaps this is an inevitable consequence of diving deep into this kind of subject matter, a kind of cipher-y Locard’s Exchange Principle where a little bit of the object’s madness brushes off onto you, however hard you try to stay aloof from it.

Regardless, the book builds up and up in a long slow crescendo towards discussing Gabor Tokai and Levente Kiraly’s (claimed) solution of the Rohonc Codex, all the way to page 130 (where Lang mentions my 2018 blog post that remained somewhat skeptical about T&K’s 2018 Cryptologia article), where… the whole thing basically stops dead.

It should be no surprise that I found this unbelievably frustrating. In football terms, he played a perfect passing sequence to get in front of an open goal, but then chose to stand on the ball. I felt like a Brazilian commentator screaming at Lang to just knock it in, KNOCK IT IN: but instead he just stood there… and then the final whistle blew.

Whereas previously I described Tokai and Kiraly’s 2018 article as a game of two halves (i.e. their codicology and block analysis was exemplary, but everything they tried to build on top of that felt a bit like a house of cards), Lang’s book feels more like just a first half. He comes across as almost in awe of Tokai and Kiraly’s work (e.g. he mentions on p.125 that Tokai has all but memorized the Rohonc Codex’s 450 pages, memorized it, I tells ya); and yet seems oddly unable to explain in print exactly what it is about their work he is so convinced by.

For me, one really epic diagram (fig. 23, p.126) taken from Kiraly’s (2011/2012) paper in Theologiai Szemle 54 exemplifies both the best and the most frustrating aspects of Lang’s book. This is because it highlights the textual wrapper that Kiraly used to infer the presence of a number system; yet also demonstrates the shortcomings of that same inferred number system but in a tiny font that is just about at the print’s limit of readability.

Essentially, that was the point that I desperately hoped Benedek would unpack the Rohonc’s claimed number system (in many ways this is a key technical aspect of Tokai and Kiraly’s work, because numbers are often an exploitable weakness of cipher systems), to make it all more tangible and understandable to his own readers (including me).

However, I can’t shift the nagging suspicion that Lang shares many of the same reservations that I had back in 2018 (e.g. his discussion of problems with the text on p.129 is very much in the same vein), but that he didn’t want to rock the boat by being negative about such outstanding guys as Tokai and Kiraly. All the same, bracketing contentious issues doesn’t actually make them go away, and if anything doing so in a book does one’s readers a disservice.

As far as it goes, then, this is a great little book on the Rohonc Codex which I’m happy to recommend for every cipher bookshelf: but quite why Lang didn’t tap the ball over the goal line still remains a mystery to me.

Somerton Man Exhumation: And So It Begins…

by ⋅ 108 Comments

As mentioned in a fair bit of Australia’s press, the exhumation of the Somerton Man has begun in West Terrace in the last 24 hours, with all the normal shots of PPE, tents, and mini-diggers accompanying the reports.

Once the FSSA have processed the body (in whatever parlous state it’s in) and extracted the man’s full DNA profile, the forensic investigation will doubtless continue spinning along for several weeks (e.g. carrying out physical analyses to determine the cause of his death), while the genealogists kick into the kind of high-octane action you’d expect. (I confidently predict “Tamam Shud: The Movie” will cast Vin Diesel as one of the genealogy team, you heard it here first.)

Then, with his identity established (hopefully), the properly fun part will begin: working out what was going on in November 1948, and fitting this with the pieces of the puzzle we know already into a full historical reconstruction.

Of course, the ultimate Cipher Mysteries prize in this whole endeavour would be a decryption of the mysterious note found imprinted on the back of the W&T Rubaiyat that was (believed to be) linked to the dead man. My suspicion, however, is that even knowing everything there is to know about the Somerton Man may still not make this possible.

Still, it’s clear that interesting times are now upon us, particularly for those people who have been promoting nutty Somerton Man theories for so many years. Perhaps we will even see some of them ‘upgrade’ their theories into denialist body-swap theories, i.e. “it’s the right DNA but the wrong body”. Praise the Lord, and pass the ammunition…

Convicts escaped from Nouvelle-Caledonie…

by ⋅ 15 Comments

While looking for lists of people “condamne a mort” in the tumult of 1871 France, I stumbled upon a list of 34 convicts who had escaped from France’s Pacific prisons in Nouvelle-Caledonie prior to 1876. (Which was what I had actually hoped to find, but hadn’t believed my archival luck was strong enough to do so.)

Anyway, the table I found in the Journal officiel de la Republique francaise, 17 Janvier 1876, p.19 (thanks to retronews.fr) was a pretty good starting point. However, according to this 2010 article by Pierre-Henri Zaidman, it was incomplete. For example, in January 1872, the minister for overseas affairs wrote that “jusqu’à présent trois évasions seulement ont été accomplies avec succès“, and Zaidman has no names for those three. (Though I’m guessing these were Villin, Patras, and Marsay.)

Even though I started by using (paywalled) filae.com to find individual bagnard’s records, I also subsequently found a (free) online database (courtesy of the Archive Nationales d’outre-mer) that allows you to search the same records (e.g. by searching the ‘Notes’ field for “evade”). This was very helpful, and allowed me to extend the search backwards by a few years.

Finally, I also found “L’ Archipel des forçats: Histoire du bagne de Nouvelle-Calédonie (1863-1931)” by Louis-José Barbançon, which is an excellent resource.

The List of Escaped Convicts

Barbançon says (p.202) that 25 convicts were thought to have escaped during 1866-1870, and 184 during 1871-1880, though I believe these figures includes les disparus.

All the same, I should point out that the length of the list below is perhaps slightly deceptive. For example, it seems certain that the entire group with Dr Paul Rastoul on 12 Mar 1875 drowned when their boat hit the reefs off l’île d’Ouen. So, while they did technically escape, it can hardly be said that they got away. 🙁

Barbançon discusses this at some length: had these convicts escaped or merely, ummm, disappeared? The reaction of the prison authorities seems to have been little more than a collective Gallic shrug: either way, such people were no longer their responsibility.

  • — 07 Jan 1866 —
  • Louis Charles Benoni Villin (no prison record, but he seems to have subsequently married Marie Damariste Phalenie Bouguignon on 06 Mar 1883 in Freniches)
  • — 12 May 1867 —
  • Etienne Lonjarret (b. 19 Jul 1833)
  • Auguste Alexandre Gence (b. 29 Sep 1837) (but appears to have died in Paris in 1887?)
  • Francois Manipoud (b. 1829), fratricide
  • Francois Marion (Marion’s body was the only one of the four that was found)
  • — 17 Jul 1867 —
  • Joseph Patras (b. 19 Jul 1843), murder
  • — 27 Aug 1869 —
  • Pierre Marsay (b. 22 Aug 1829)
  • — 6 May 1873 —
  • Isidore Petit (b. 12 Oct 1840)
  • — 9 Nov 1873 —
  • Jules Deslandes, 29 years old, “tourneur-repousseur” (and Communard)
  • — 3 Jan 1874 —
  • Edmond Moriceau (b. 25 Mar 1837), the notes say he was supposed to have “parti pour Sydney”, but also that he died 4 May 1879?
  • — 7 Jan 1874 —
  • Paul Robin 1837-1912
  • — 27 Jan 1874 —
  • Two (unnamed?) convicts escaped (according to Zaidman)
  • — 20 Mar 1874 —
  • Paschal Jean Francois Grousset 1844-1909 [politician, journalist, translator and science fiction writer] wrote “Les condamnés politiques en Nouvelle-Calédonie” (1876) with Francois Jourde
  • Olivier Pain 1845-1884 [journalist]
  • Victor Henri Rochefort de Lucay 1831-1913 [Marquis de Rochefort-Luçay, see “Henri Rochefort : déportation et évasion d’un polémiste” (2004) Joël Dauphiné]
  • Francois Jourde 1843-1893 (wrote “Souvenirs D’Un Membre de La Commune“)
  • Achille Ballière 1840-1905 (architect, wrote “La Déportation de 1871: Souvenirs d’un Évadé de Nouméa“)
  • Charles Bastien
  • Charles Grantille (perhaps Grandthille?)
  • — 23 May 1874 —
  • Francois Coutouby, 37 years old, “agent d’affaires et marchand de vin”
  • — 20 Jan 1875 —
  • Ernest Harenger, 37 years old, “cordonnier, ancien militaire” (believed drowned during an attempted escape)
  • — 12 Mar 1875 —
  • Eugène Barthélemy (b. 17 Sep 1847)
  • Martin Louis Berger (b. 12 Oct 1841)
  • François Palma (b. 02 Jun 1840)
  • Michel Eugene Galut (b. 09 Mar 1841)
  • Vincent Guigue (FR ANOM COL H 84)
  • Alexandre Eugene Gilbert (FR ANOM COL H 82)
  • Charles Auguste Emile Demoulin (b. 30 Aug 1851)
  • Pierre-Marie Alexandre Masson (b. 31 Jan 1847)
  • Mathieu Chabrouty (b. 13 May 1853)
  • Marcel Julien Roussel (b. 01 Apr 1850)
  • Louis Auguste Leblant (b. 30 Jan 1838)
  • Henri Gaston Edat (b. 21 Feb 1854)
  • Louis Garnier [no mention of an escapee by this name in the prison files, though convict Louis Hubert Garnier died in hospital in 1875?]
  • Jean Savy (b. 15 Sep 1838)
  • Dr Paul Emile Bethelemy Philemon Rastoul (b. 01 Oct 1835)
  • Auguste Ledru (b. 22 Jun 1829)
  • Jean Antoine Auguste Saurel (b. 06 Dec 1842)
  • HippoIyte Jules Sauvé (b. 07 Sep 1839)
  • Prosper Gaspard Ephege Adam (b. 16 Jan 1848)
  • Edouard Duchesne (b. 06 Dec 1842)
  • — 20 Jun 1875 — (group landed at Wide Bay near Maryborough, all given an amnesty in 1879)
  • Emile Charles Paty (b. 16 Nov 1842)
  • François Décombes (b. 11 Mar 1833)
  • Laurent Brissard (b. 09 Jul 1845)
  • Pierre Graillot (b. 08 Jan 1851)
  • Alexandre Joseph Rousseau (b. 02 Jul 1841)
  • — 4 Jul 1875 —
  • Louis Jean Baptiste Merchez (b. 04 Mar 1842) [Note that he appears to have had a son Paul Henri Merchez in 1886 with his wife Zaire Irma Hennion (b.1846)]
  • Eugene Sellier (Aged 37 in 1874)
  • — 10 Jul 1875 —
  • Gilles Etienne Excoffier (b. 13 Nov 1843), journalier, house-breaker [Appears to have died in 1917]
  • — 27 Oct 1875 —
  • Claude Faury (b. Jan 1843)
  • — 9 Nov 1875 —
  • Adolphe Eugene Fabret (aged 41 in 1874)
  • Jevin (?)
  • Denis Louis Roch Siblanc (aged 29 in 1873))
  • Martin (?)
  • Barrely (?)

More for your Manet…

Finally, just because I like to spoil you, here’s Manet’s painting of Henri Rochefort and his five fellow Communard escapees rowing from Nouvelle-Caledonie to Australia.

Final notes on Jean Keff and Pierre Keff…

by ⋅ 4 Comments

Before moving on, I thought I ought to publish my last few notes on Jean Keff and Pierre Keff, in case someone passing happens to be trying to work out their family tree. (I should again stress what a good source of information filae.com is.)

Children of Jean Keff & Reine Lichtenberger

Note that in the notes for Jules Joseph Keff, Jean Keff is referred to as “Jean Pierre Keff”.

Leon Pierre Jean Keff
b. 20 Feb 1865, Paris
d. 27 Feb 1900, Paris (named as "Pierre Jean Leon Keff")

Pierre Jean Baptiste Keff
b. 20 Jun 1866, Paris

Anais Keff
b. 24 Feb 1869, Paris
d. 06 May 1887, Paris

Jules Joseph Keff
d. 9th Match 1869, aged 2 years 9 months

Children of Pierre Keff & Catherine Birschens

Note that Pierre Keff’s profession is given as both “tailleur” and “polisseur”, while Catherine Birschens is referred to as “cartonniere”.

Catherine Birschens died on 13 Mar 1871 at Epinay-sur-Orge.

Victorine Josephine Keff
b. 27 Nov 1864
d. 03 May 1865 at 4 Rue Grange aux Belles, aged 5 months

Josephine Catherine Keff
b. 21 Jun 1863, Paris
m. (no date) Victor Pierre van de Casteele
d. 14 Mar 1944, Paris

Sadly, Henry Debosnys was not Pierre Keff… :-(

by ⋅ Leave a Comment

I was certain the French archives would have good, solid information on historical prisoners, so went looking for them. And that’s how I found the official French archive site filae.com, which (modest paywall notwithstanding) was actually very impressive (To be precise, searching filae is a bit hit and miss, but the depth of the archives is excellent.)

So I can now say with certainty what happened to Jean Keff and Pierre Keff (oddly, I wasn’t able to find prison records for Marie Ratier, the widow Bon).

What Happened to Jean Keff?

As we already know, Jean Keff was given a life sentence in 1872 for the attempted rape of minor Henriette X, and his appeal was rejected. He then escaped from Le Bagne de Toulon on 14 Oct 1872, but was recaptured on 22 Oct 1872.

His prison records take up his story. (There is a note that he had a previous 13-month conviction in 1860.) For attempting to escape, he was sentenced to three years in ‘double chains’. He was then unchained on 25 Jan 1873 and transferred to Nouvelle-Caledonie on the transport ship Le Rhin. There, his attempts to escape continued:

  • 20 Oct 1873: escaped Ile Nou, recaptured 22 Oct 1873
  • 23 Dec 1875: escaped Ile Nou, recaptured 25 Dec 1875
  • 1879: escaped Ile Nou, recaptured
  • 4 May 1880: escaped Ile Nou, recaptured on the 6th
  • Jul 1884: not sure what happened here

Sadly, the final entry in Jean Keff’s three pages of prison records isn’t hard to predict:

  • Died at Ile Nou, 10 Feb 1894

What Happened To Pierre Keff?

From the reports we have already seen, we know that Pierre Keff too was given a life sentence in 1872 for his part in the conspiracy to rape Henriette X. His appeal against the sentence was upheld, but he was due to be rearrested and retried.

From his prison record, we can see that that the sentence given to him in his retrial was 15 years: he arrived at Le Bagne de Toulon on 09 Sep 1872. As with thousands of other bagnards, he was subsequently transported to Nouvelle-Caledonie on 18 Apr 1873. Despite escaping from Ile Nou (on 02 Dec 1873), he was recaptured on 14 Dec 1873.

However, when the authorities responded (on 06 May 1874) by extending his sentence by two years, that punishment seems to have put him off trying to escape again, because – unlike his brother Jean’s long list of escape attempts – the next event in Pierre Keff’s prison record is his release on 11 Aug 1889, 17 years after his trial.

A search of filae.com’s death records found the same Pierre Keff (still born in Chateau-Rouge to Pierre Keff and Catherine Andre, but now a widower of Catherine Derichelle) dying in Paris at 11.45am on 15 Jun 1911. His brother Jean is also mentioned here as having died in Ile Nou.

So I think there is, alas, no way that Pierre Keff could have been Henry Debosnys.

So… Where Next?

Given that I mistrust just about everything wife-killer Henry Debosnys claimed as his history; that his body had what looked like prison tattoos (the Philadelphia Times noted that Debosnys seemed like a French convict); and that the French journalist claiming to have known him thought his real surname was Keff and that he was born around 1833, it wasn’t entirely unreasonable to wonder whether Pierre Keff – born 1833, and sent to Le Bagne de Toulon for conspiracy to rape a minor – might have been both people.

Intriguingly, Stefano Guidoni notes that Debosnys’ Portugibberish notes included the word “noumea”, which is the capital of Nouvelle-Caledonie. So, much as I doubt both the veracity and the sincerity of all Debosnys’ writings, there remains a vague suspicion that there may be something real peeking through the cracks there.

But was this even the same M. Keff described by the Parisian journalist, “a good-sized fellow with long black hair, a smooth, fat always carefully shaved face“? Here’s his physical description when he left prison:

The blond beard sounds somewhat inconsistent, hein? But it is Pierre Keff’s profession – “polisseur” – that is arguably the most inconsistent with the journalist’s account.

I don’t know: right now, I doubt I’ve even found the right Keff, never mind the right Debosnys. And for Keffs, the French historical prison records list only the two brothers.

Perhaps the historical records of Charlemagne college will be able to throw a little more light on this, if (as claimed) Keff attended there in 1845 as a 12 year old. At this distance in time, this might be the only practical way to verify the journalist’s story. (Did he even have Keff’s surname correct? We don’t know.)

At the same time, I’m wondering whether it might be worth looking at escaped French convicts from the period 1870-1880. I mentioned a few before, but Docteur Raoulx might possibly have included a list in his 1929 book. Something to think about, anyway…