Category Archives: Historical Ciphers

I’ve just watched a sort-of-mostly-interesting voynich.ninja interview with Adam Lewis, whose senior thesis at Puget Sound I mentioned here a few days ago.

However, one particular part of his argument (as presented in the interview, at least) made no real sense to me at all: that even though the Voynich offers (Lewis says) a decent-sized corpus of text to work with [in fact, it’s possibly the largest ciphertext from the pre-machine-cipher era], its text style is somehow too ‘samey’ for linguists to find the weak link in (and hence crack).

In fact, we have (at least) two ‘languages’ to work with, better known in Voynich circles as Currier A and Currier B (named after the distinguished US code-breaker Captain Prescott Currier, who first noted them in the 1970s), though it remains an open question as to whether or not they are closer to ‘dialects’ than distinct ‘languages’ per se. Hence I’m not entirely convinced by Lewis’s argument.

Regardless, might there be some subtle thread (whether linguistic, cryptological, or whatever) connecting two different blocks of text that we can trace between the two pages they appear on? Might we be able to find a tentative internal Voynichese text match (be it ever so small) to cast some faint light on both parts, in precisely the way that Adam Lewis asserts that we cannot?

Well… I hope you’ll agree that it’s worth attempting, so let’s give it a go, eh? 🙂

Three Pages, One Plant?

I believe that most Voynich researchers would agree that – very unusually – a single plant seems to appear in three separate places in the manuscript: f17v, f96v, and f99r. Long-suffering Cipher Mysteries readers may also remember that I posted about these pages back in 2013.

Note that identifications of the plant are somewhat nebulous and unsure:
* Dana Scott – wild buckwheat (Ellie Velinska tends to agree with this, or perhaps bryony)
* “biologist from Finland” – – Rumex acetosella? or Smilax aspera (=common smilax) or Smilax excelsa. Or Dioscorea communis / Tamus communis?
* Peter – Tamus communis or Smilax aspera
* MarcoP – Cadamosto’s “vitis nigra” (black vine) commonly “tamo” (“vitis nigra tamo dice lo vulgo”), i.e. Tamus communis
* JKP – suggests many different plant matches
* Kieran Coughlan – Creeping Nightshade (Solanum Dulcamara)
* Edith Sherwood – yam (even though yam is post-Columbus) – and she thinks she can see Leonardo da Vinci’s signature hidden in its hairy roots. As always, make of that what you will.

(Doubtless there are many more to be found, but that’s the point where I lost interest.)

Codicologically: even though f96v and f99r would at first appear to have had a pair of pages inserted between them at some stage (i.e. f97 and f98), they ended up facing each other in the Voynich’s final state. Moreover, f96v is the last page of quire 17, and f99r is the first page of quire 19: my best reading (that I put forward in 2012 in Frascati) is that there was never any such pair of inserted pages, but that instead someone else had previously misnumbered the final two quires (Quire 19 and Quire 20), and that the two extra pages there were added to account for the ‘Quire 18’ that was never actually there. Perhaps none of this is actually important: but I think there’s a good chance that – contrary to their final folio numbering – f96v and f99r have sat facing each other for some time, perhaps even in the original gathering order.

Anyway, comparing the blocks of text reveals something I think is rather unusual, and which may possibly come as a surprise to many who look more at the drawings than at the text: that even though they share the same plant drawing, they really are – when you look a little closer – very different indeed from each other.

f17v

1. pchodol.chor.fchy.opydaiin.odaldy
2. ycheey.keeor.cthodal.okol.odaiin.okal
3. oldaim.odaiin.okal.oldaiin.chockhol.olol
4. kchor.fchol.cphol.olcheol.okeeey
5. ychol.chol.dolcheey.tchol.dar.ckhy
6. ockhor.or.okaiin.or.otaiind
7. sor.chkeey.paiir.cheor.os.s.aiin
8. qokeey.kchar.ol.dy.choldaiin.sy
9. lcheol.shol.kchol.choltaiin.ol
10. oytor.okeor.okar.okol.daiir.am
11. qokchey.qokaiir.ctheol.chol
12. oy.choy.kaiin.chckhey.ol.chor
13. ykeor.chol.chol.cthol.chkor.sheol
14. olor.okeeol.chodaiin.okeol.tchory
15. ychor.cthy.chshky.cheo.otor.oteol
16. okcheol.chol.okeol.cthol.otcheolom
17. qoain.shar.she.dol.qopchaiin.cthor
18. otor.cheeor.ol.chol.dor.chr.oreees
19. dain.chey.qoaiin.cthor.cholchom
20. ykeey.okeey.cheor.chol.sho.odaiin
21. oal.sheor.sholor.orshecthy.cpheor.daiin
22. qokeee.dar.chey.keeor.cheeol.ctheey.cthy
23. chkeey.okeor.shar.okeom

A list of mildly distinctive features of f17v might well include:
* the “ol.olol” repetition on line 3
* the single-leg gallows on line 1 (possibly a Neal Key?) and on line 4. (Might line 4 in fact be the start of a second paragraph?)
* note that the single-leg gallows on line 17 also looks like it might be the top line of a paragraph
* the “qoain” on line 17 (this appears only 7 times in the manuscript) and “qoaiin” on line 19 (this appears only 23 times)
* the “autocopied”-like choldaiin/choltaiin pair on lines 8 and 9
* chol appears 11 times (as itself and within other words)
* chol.chol (line 13) appears 39 times in A pages but only a single time in B pages

f96v

1. psheossheeor.qoepsheody.odar.ocpheeo.opar.ysarorom
2. yteor.yteor.olcheey.dteodaiin.sary.qoches.ycheom
3. dcheoteos.cpheos.sor.chcthory.cth.ytchey.daiin
4. dsheor.sheey.teocthy.ctheodody
5. tockhy.cthey.ckheeody.ar.cheykey
6. yteeody.teodar.olchey.sy
7. sheodal.chorory.cthol
8. ycheey.ckheal.daiin.s
9. oeol.ckheor.cheor.aiin
10. ctheor.oral.chor.ckhey
11. sar.os.checkhey.socthh
12. sosar.cheekeo.dain
13. soy.sar.cheor

Here:
* the four single-leg gallows (p) on line 1 could very easily be flagging Neal Keys in some way (e.g. [p] + “sheossheeor.qoe” + [p])
* there are only three instances of “yteor” in the whole manuscript, and line 2 has two side by side
* the way that “s” is used as the first character of lines 11-13 gives the appearance of being filler in some way (e.g. a null)
* there’s also an autocopyist-style diagonal set of “sar”s on lines 11-13 (sar / sosar / soy.sar) that also looks like filler
* the repetition of ctheodody and ckheeody on lines 4 and 5 makes it look as though ctheodody may have been a miscopying of ctheody

Hence even though the use of ‘language’ superficially appears similar to f17v (Currier-wise), there are very few actual similarities.

f99r, final paragraph

1. tolkeey.ctheey
2. ykeol.okeol.ockhey.chol.cheodal.okeor.olcheem.orar
3. okeeey.keey.keeor.okeey.daiin.okeols.aiin.olaiir.oolsal
4. qokeey.okeey.qokeey.okesy.qokeey.sar.sheseky.or.al
5. yshain.yckhey.octhey.dy.daiin.okor.okeey.shcthy.sh
6. ychor.ols.or.am.airam

Here, “qokeey.okeey.qokeey.okesy.qokeey” on line 4 is exactly the sort of repetitive garbage that reduces attempted linguistic decryptions to mush while simultaneously giving heart to hoax theorists and their CompSci tables.

A vs B, again

We can try categorize these three pages by revisiting Currier’s original notes on A and B,:

(a) Final ‘dy’ is very high in Language ‘B’; almost non-existent in Language ‘A.’
(b) The symbol groups ‘chol’ and ‘chor’ are very high in ‘A’ and often occur repeated; low in ‘B’.
(c) The symbol groups ‘chain’ and ‘chaiin’ rarely occur in ‘B’; medium frequency in ‘A.’
(d) Initial ‘chot’ high in ‘A’; rare in ‘B.’
(e) Initial ‘cth’ very high in ‘A’; very low in ‘B.’
(f) ‘Unattached’ finals scattered throughout Language ‘B’ texts in considerable profusion; generally much less noticeable in Language ‘A.’

Additional observations noted by Rene Zandbergen:

(g) The very frequent character combination ed is almost entirely non-existent in all A-language pages.
(h) The very common character combination qo is almost completely absent in the zodiac pages and the rosettes page, but appears everywhere else.
(i) The common character combination cho does not appear in the biological pages (and the rosettes page), but it does in other B-language pages.

From this, we can say (somewhat uncomfortably) that:
* f17v is a definitely-Currier-A page (it has lots of “chol” and “chor” instances, as well as cth- words)
* f96v is a sort-of-a-Currier-A page
* f99r is a sort-of-a-Currier-B page

And so we get to the awkward situation I highlighted back in 2013, that A vs B is far too simplistic to be a globally useful razor, and that we also don’t yet have a good ‘roadmap’ for how the ‘language’ used in the Voynich Manuscript evolves and changes through its pages. Which is essentially why I think that Adam Lewis is being too reductive when he claims that Voynichese is too samey across its pages to be crackable – the actual situation would seem to be that Voynichese is actually too unsamey for anyone to get a grip on.

But is there a match?

Back in 2013, commenter Šuruppag suggested on my original page that there might possibly be a very short match between all three blocks of text. Specifically (as transcribed above):

* f17v line 5: ychol.chol.dolcheey
* f96v line 2: yteor.yteor.olcheey
* f99r line 3: keey.keeor.okeey

Šuruppag continued:

A nice little sequence that occurs once in each passage. Seems to show similar pattern and characters.
Perhaps this sequence (possibly the name of the plant?) shows us the evolution of the cipher. Feel free to criticize if you don’t think so.

I would add that there’s reasonably good evidence (I flagged this in The Curse of the Voynich in 2006) that “-y” may mark where a longer word has been truncated: and so “keey” on f99r could very easily be a (slightly) truncated version of “keeor”. In which case we might wonder whether “yt” on f96v is enciphering the same underlying letter as “k” on f99r.

Again, in Curse in 2006 I proposed that EVA ok / ot / yk / yt (all of which appear prominently in label-driven sections such as the zodiac pages, e.g. “otolal”) might well be verbose cipher pairs, whereas EVA t / k (i.e. where not immediately preceded by o or y) might well be some kind of transposition cipher where a different letter gets inserted from elsewhere in the text (t and k are basically interchangeable in almost all circumstances). This makes me suspect here that the last word of the phrase on f99r should actually be olkeey rather than okeey, because “ok” would then be a verbose cipher whereas “olk” would be parsed as “ol-k”.

I also wrote in Curse that I thought not only that EVA e / ee / eee / ch shapes were verbosely enciphered vowels, but that their mapping and use seemd to change between A and B pages. So I for one would not be hugely disturbed if it were to be the case that EVA “e-or” in f96v reappears as EVA “ee-or” in f99r.

And finally, I have also proposed that where Voynichese words start with “d-“, this may well turn out to be a signal that a longer word has been split up into two shorter words. (For example, I suspect that daiin daiin sequences encipher groups of Arabic digits that should be joined together.)

Putting all these fragments together, I wonder if perhaps the three lines (if the various miscopyings were corrected, and the f99r -y rectified as -or, and the words reassembled) were originally supposed to be parsed as:

* f17v line 5: ychol..chololcheey – – – or perhaps “ychol.ychololcheey”
* f96v line 2: yteor.yteorolcheey
* f99r line 3: keeor.keeorolkeey

Hence it seems to me that what might well be going on in the f99r line is that three “k”s are being used to stand in for yt / yt / ch respectively, i.e. that the encipherer is using a transposition (i.e. replacement from elsewhere in the text) cipher trick rather than f96v’s verbose cipher trick to encipher the same plaintext word.

But what then of f17v’s text? My best current guess is that this version may well be closer still to the same small block of plaintext being enciphered on all three pages.

One cipher alphabet, multiple verbose ciphers?

More awkwardly than all of the above, the above seems to suggest that chol in the first page be the same as eor in the second page and eeor in the third page: that is, that even though the Voynichese letters are the same across different pages, the arrangement of verbose ciphers using those covertext letters may well be changing. Is our inability to read Voynichese then largely a consequence of multiple verbose cipher arrangements having been used?

When this kind of Voynichese discussion comes up, I often think back to my late (and intensely argumentative) friend Glen Claston. When he was building up his Voynichese transcription, he often noticed (as he mentioned to me a couple of times) that the patterns and styles of words would “shift” every few lines or paragraphs, as though it internally changed into a different gear or mode. He was never able to quantify this shift more precisely (even though he was a keener observer than almost everyone), but perhaps there lurks at the heart of Voynichese some kind of verbose cipher modality – not enough to disrupt the overall “look” (i.e. of the covertext), but more than enough to throw the decryptor hounds off the crypto-scent.

That is, perhaps what these three lines is trying to tell us is that we should be looking not for a single global verbose cipher key (as might be suggested by the way that the covertext consistently uses a relatively small set of cipher glyphs), but for a number of different (and rather more localized) verbose cipher keys. Voynichese’s covertext may be formed from a single set of glyphs across all its pages, but might the complexity around the different blur of Currier ‘languages’ actually be telling us that there are multiple encryption schemes in use?

If that is correct (to even some degree), what would we need to do to get under Voynichese’s skin? I’ve often spoken of the need to try to map Currier A usage to Currier B usage (to try to identify equivalent sequences in the two parts), but perhaps this is only 5% of the larger challenge, and moroever a step that makes no sense until the verbose ciphers have been even partially mapped – for it could very well be that a part of a single page might well need to be remapped to a different part of the same page in order to be able to parse it, let alone read it. Perhaps the first step should be to try to find examples of Voynich Manuscript pages that exhibit Clastonian multimodality (i.e. where there is an apparent shift in the system within the page), and see if we can quantify how the changes in behaviour work in practice, just so that we can even try to parse what is going on.

Tricky, though. 🙁

Incidental news: Cipher Mysteries recently had its 25,000th comment, while it has now also had close to three million page views. Which is nice. (Just thought I’d mention it in passing.) The remainder of the post is for tying up various Voynich threads that aren’t each enough for a whole post of their own.

Edith Zimmerman

Here’s a Voynich amuse-bouche: any page with drawings taken from the Voynich Manuscript’s Quire 13 that includes the following quote is more than OK by me:

Anyway, so then we got to the part in our performance where Shelly and Mathilde stand in giant pipes covered with rhinestones. They hold up large balloons that have fishtails dangling off them, Chinese dragon-style.

Voynich LOL

An example of the per-section language use in Voynichese is that the word EVA “lol” appears far more often in quires Q13 and Q20 than anywhere else. Here’s the lol cluster in the top paragraph of f77r (1 x loly, 3 x lol, all at the end of lines):

Aside from Q13, EVA lol also appears here: f48v f86v6 f103r/v f106r f107r/v f111v f113v f115r f116r. Incidentally, the first word of the last paragraph of f116r (which I suspect contains a colophon from the author) seems to me to be the same kind of thing as the heavily digraph-structured labelese words, though with the extra twist of EVA l sometimes standing in for ol, as I discussed elsewhere):

Brian Hendley

The recent talk of Canadian academics’ rogue AI’s preparing for the enslavement of Mankind by solving Voynichese triggered some memories in the head of Brian Hendley, a philosophy professor at the University of Waterloo:

Your recent story about deciphering the mysterious Voynich manuscript took me back to my graduate school days at Yale and my dissertation supervisor, Bob Brumbaugh. Bob was a recognized Plato scholar but he did have a crack at deciphering the Voynich manuscript, resulting in some published articles and a book, “The Most Mysterious Manuscript: The Voynich ‘Roger Bacon’ Cipher Manuscript.” I recall Bob saying that you could spend your entire scholarly career trying to decipher the Voynich.

Ain’t that the truth, eh? *sigh*

Gerard Cheshire (yes, again)

Polyglot linguistic Voynich theorist Gerard Cheshire (whose theory I discussed here back in 2017) has hurtled into 2018 even more convinced of his utter rightness (and of everyone else’s abject wrongheadedness).

If you really, really want to read his all-new polyglot interpretation of the nine rosette page (which he calls “Tabula regio novem”, which I don’t believe is grammatical Latin, whatever Google Translate may say) which is all about Italian volcanoes, it is online here.

For those whose appetite for such things would be easily assuaged by a single dim sum, here is Cheshire’s elucidation (I hesitate to call it a decryption or translation) of the text around the bottom-left (SW) rosette:

om é naus (people and ship)
o’monas (in unity)
o’menas (take charge)
omas (mothers/babies)
o’naus (of ship)
orlaus [orlas] (to protect)
omr v asaæe [vasaie] (life-force pots: pregnant bellies)
or as (yet in)
a ele/elle a (he/she at)
a inaus [inauspitica] (inauspicious/unfavourable),
o ele e na (he/she is in)
æina (a/one)
omina (omen)
olinar (to look)
n os aus (it is)
omo na moos (man not mouse)
é ep [epousee] as (and embrace)
or e ele a opénas (an opening)
os as ar vas (thus you go)
opas (but carefully)
a réina (to the queen)
ol ar sa os aquar aisu na (to facilitate not getting wet with seawater).

J. Michael Herrmann

One little-noted Voynich theory from last year was J. Michael Herrmann’s The Voynich Manuscript is Written in Natural Language: The Pahlavi Hypothesis. Exactly as it says on the tin, it’s a linguistic Voynich theory:

Here, we provide evidence that the VM is written in natural language by establishing a relation of the Voynich alphabet and the Iranian Pahlavi script. Many of the Voynich characters are upside-down versions of their Pahlavi counterparts, which may be an effect of different writing directions. Other Voynich letters can be explained as ligatures or departures from Pahlavi with the intent to cope with known problems due to the stupendous ambiguity of Pahlavi text.

Herrmann followed that with his 2018 paper The Cannabis Page of the Voynich Manuscript. This finesses some of the claims in the 2017 paper, now suggesting that Voynichese uses an “alphabet that has similarities to Pahlavi and Mandaic script”.

The “Cannabis Page” referred to in the title is f16r:

In Herrmann’s translation (which is accompanied by extensive notes, presumably to help readers overcome the “stupendous ambiguity” of Pahlavi), the first paragraph of f16r emerges as a tirade against the evils of cannabis:

Cannabis [is] vain. Stay away from the impudent crowing man. The pipe is a debasement. Jaundice [is] the overly happy face of the adherent. [He is] puffed up with pride. Security does not come [to him]. In the evening peace of mind does not come [to him, as] the serpent of nightly lust spoils him. [What is] concealed,
will become public.

If you are interested, Herrmann’s rendering of the first paragraph of f1r is as follows:

(1) The humble grass shames you. Obediently hold the law. You cry for help for the tribunal commanded (2) for being insolent. Verily, you are frightened out of your wits by the troops in fury. Crowds and crowds of abled ones. (3) Woe, the well ordered line (of troops) is driving forward. If you see this, you will. Don’t start to count and to number the men, woe! (4) Refrain form [sic] the uneducated baldhead, the storyteller, the teaching of the “light bringer”. Remember the time of the fathers. (5) Heaven shield those who are weak for debasement. The man of doubt guide right.

If only I had an open letter I could use, if only… :-/

Thanks to Newsweek, Fox News, The Daily Mail and The Independent [*sigh*], some techy Canadian Voynich research is currently enjoying its day in the media sun. (Hint to authors: sorry, but based on recent evidence, it would seem that you have ~48 hours to get your next funding request submitted and approved before everyone currently cheering starts booing.)

CompSci professor Greg Kondrak and graduate student Bradley Hauer presented their research at the 2017 ACL conference, and their paper “Decoding Anagrammed Texts Written in an Unknown Language and Script” appeared in Transactions of the Association for Computational Linguistics Volume 4, Issue 1, pages 75–86 [though the PDF is freely downloadable, at least for now].

From the press coverage so far, you might think that they had CARMELed the Voynich (i.e. thrown a tame supercomputer and some kind clever-arse AI libraries at the problem): for, as the media incessantly repeat at the moment, All Human Problems Will Inevitably Yield To The Scythed Mega-Bulldozer That Is AI. But… is any of that true? Or useful? What’s actually going on here?

Behind the Kondrak and Hauer headlines

The initial question is obvious: what did Kondrak and Hauer actually do to try to crack the Voynich’s mysterious secrets that (they thought) nobody else had tried before? A quick snoop reveals that Bradley Hauer is a pretty smart crypto cookie: the simple substitution cipher solver presented in his 2014 paper “Solving Substitution Ciphers with Combined Language Models” outperforms many competing academic solutions. It does this by using both letter statistics and word lists at the same time (a) to solve Aristocrat cryptograms (i.e. ones where you know where the word boundaries are) even under mildly noisy conditions, and (b) to solve Patristocrat cryptograms (i.e. ciphertexts without spaces, though the recursive approach used to turn Patristocrats into candidate Aristocrats seems somewhat heavy-handed), before finally moving on to trying (unsuccessfully) to reproduce the kind of deniable encryption loosely proposed in Stanislaw Lem’s (1973) “Memoirs found in a bathtub”.

And here’s what Hauer looks like in real life:

So what happened before the Voynich paper was even written was that Hauer had built up a lot of software machinery for solving nicely-word-boundaried simple substitution ciphers at speed, and where some kind of mild text mangling had optionally taken place. And so it should not be a surprise that he carried this technology and approach forward, insofar as the 2016 paper tries to solve Voynichese as if it were a nicely-word-boundaried simple substitution cipher that had had its text mangled via anagramming plus optional abjad-style vowel removal. Given that as the paper’s founding presumption, all it is trying to do is evaluate which plaintext language was used if that entire presumption just happened to be correct (oh, and the transcription used was accurate).

Incidentally, the Voynich corpus used was 43 pages (“17,597 words and 95,465 characters”) of Currier-B text in the Currier transcription that one or both of Knight & Reddy had supplied, but the authors did not seem to have questioned the reliability or parsing choices behind that particular transcription. (More on this below.)

Voynich anagramming

Unlike Stephen Bax’s well-known Voynich 2014 paper (which began by gleefully flipping the bird at nearly all previous Voynich research), Kondrak and Hauer’s Voynich paper begins by covering what they consider related Voynich work (section 2.1) in a level-headed, if somewhat brief, way. The most relevant source they have for the notion that we might be looking at anagrammed text is Gordon Rugg’s 2004 paper: this floated the idea that there might be a similarity between alphabetically ordered anagrams (‘alphagrams’) and what we see in the Voynich Manuscript’s text.

Yet much has already been written about Voynich anagramming beyond this, not least William Romaine Newbold’s monstrously tangled ‘decryption’ (*shudder*). More recently, Edith Sherwood claimed both that it was a young Leonardo da Vinci who wrote the Voynich Manuscript, and that the Voynich text was written in anagrammed Italian (though so far she has mainly only tried to reconstruct Voynich plant names using her proposed scheme). As I pointed out in 2009 this seems extraordinarily unlikely to work in the way she proposes.

Arguably the most interesting previous Voynich research into anagrams (again, not mentioned by Hauer) has been that of London-based researcher and translator Philip Neal. In a (now long-lost) page he posted many years ago on the late Glen Claston’s voynichcentral.com website, Philip proposed:

Here is a transformation of plaintext into ciphertext which explains certain features of the Voynich “language”.

1. Divide a plaintext into lines
2. Sort the words of each line into alphabetical order
3. Sort the letters of each word into alphabetical order

1. one thing led to another thing last night
2. another last led night one to thing thing
3. aehnort alst del ghint eno ot ghint ghint

The result has some of the statistical properties of the Voynich text.

A. The frequency distribution of words and letters is the same as in the natural language plaintext, but the distribution of two-letter groups and two-word groups is significantly altered.
B. Words at the beginning of a ciphertext line tend to start with letters at the beginning of the alphabet. Compare the high frequency of Voynich “d” at the beginning of a line.
C. If a letter near the end of the alphabet has a tendency to be word-initial in the plaintext (e.g. German “w”), it will have a strong tendency to be the last word in a line. Compare the high frequency of Voynich “m” at the end of a line.
D. The ciphertext versions of frequent words will tend to cluster together in a line. That is, where a word such as “thing” occurs twice in the plaintext line (as in the above example) the two word sequence “ighnt ighnt” will occur, but “ighnt” may also occur elsewhere in the line as an anagram of “night”.
E. A one-letter word of ciphertext can only be an anagram of a single word of plaintext (“a” can only be an anagram of “a”) and a two-letter word of ciphertext can only be an anagram of two possible words of plaintext (“et” can only be an anagram of “et” and “te”). This means that you cannot have a ciphertext line of the pattern “… i … i … ” or of the pattern “… et … et … et …”. This principle largely holds good in the Voynich text: there are only six exceptions in the corpus of Currier’s language B.

To his credit, Philip then immediately pointed out some problems with this suggestion:

1. Voynichese words do not conform to a strict alphabetical ordering of letters (there are quite a lot of words of the pattern dshedy).
2. Voynichese words have a strong tendency to contain only one instance of a given letter, unlike any obvious candidate language for the plaintext.
3. The enciphering described is not unambiguously reversible (however I think it would work as a private aide-memoire, or as a means of establishing priority like Galileo’s well known anagram announcing his discovery of the phases of Venus)

(Philip has since instead proposed a possible grid-like constraint on the position of Voynichese letters within Voynichese ‘words’, though problems with that alternative explanation remain.)

Incidentally, Philip has also pointed to a number of places within the Voynich Manuscript where entire lines appear to have been written in a non-one-after-the-other way (i.e. unexpected line transpositions): while nobody has yet come up with a powerfully convincing explanation for the presence of “Neal keys” (sections of text typically delimited by pairs of single-leg gallows) in the top lines of pages (typically embedded ⅔ of the way across). He is a sharp observer, and these anomalies are all inconsistent with the widely-held presumption that the text we are looking at here is completely unmangled.

Ultimately, though, it remains a sizeable step (or three, or indeed more) to go from anywhere here to Hauer’s presumption that what we are looking at is straightforwardly anagrammed text in a conventional European language, whether abjad or not.

The actual Voynich research gap

If asked for the single largest methodological problem with Voynich research, I would point to the way that Voynich researchers tend to make a series of unfounded assumptions:
(a) the transcription they are using is perfectly reliable;
(b) the way that they parse that transcription (i.e. into tokens) is correct – there are many hidden linkages here which are each probably sufficient to derail any decryption attempt;
(c) the candidate plaintext languages they consider are genuinely representative of the Voynich Manuscript’s plaintext;
(d) no other textual transformations are present;
(e) the putative hypothetical transformation that they just happen to have plucked from the air and which they are testing is precisely that which is present in the Voynich Manuscript; and
(f) the output of their reverse transformation will be straightforward text that can be read and marvelled over by historians.

In the case of Kondrak and Hauer, I hope it should be clear that they have fallen foul of every one of these issues in turn: and their paper is all the worse for it. It is one thing to note in passing that Esperanto’s “extreme morphological regularity […] yields an unusual bigram character language model which fits the repetitive nature of the VMS words” (p.83), but it would be quite another to point out that this might easily have arisen from the way that Voynichese needs to be parsed in order for it to make sense: and it is this apparent lack of perception of the practical difficulties that all Voynich decryptors face that devalues the genuinely good work that went into their paper.

What particularly frustrates me is that in spite of these many issues, there are plenty of ways Voynich researchers can make genuine progress towards understanding what is going on: but, rather, they instead persist in trying to airball their own personal Voynich match-winner from the other end of the basketball court. They seem seduced by the glamour of being The One Who Solved The Voynich, instead of getting on with the graft of making a difference to what we know. 🙁

Yet computational linguistics has such a rich toolbox (of which CARMEL is merely one small screwdriver) that it surely has ample capacity to at least try to bridge all the actual research gaps that people are falling into, e.g.:

* What is the right way to parse EVA into tokens? (e.g. is EVA ‘or’ two tokens or one? is EVA ‘cth’ three tokens, two tokens, or one? etc)
* How does Currier A map to Currier B? And what about all the subtypes of each of these?
* What are the differences between them and “Currier C”? (Rene Zandbergen’s term for labelese)
* Can we determine whether line-initial letters are likely reliable or unreliable?
* Are words abbreviated (e.g. is EVA y some kind of truncation symbol)? If so, are A and B abbreviated in exactly the same way?
* etc

If people had the intellectual good sense to stop trying to fly over all these separate hurdles all at the same time in a Steve Austin-style 100m leap of misplaced faith, we might start to make real progress. However, even when researchers do have the necessary brains to make progress (as Hauer clearly has), it seems they have insufficient strength of mind to not be tempted by the glamour of the big ticket “Researchers Crack Voynich Manuscript” headline. 🙁

Because last year’s Voynich research brought me a step closer to German calendars, I finally got round to reading Ernst Zinner’s epic “Regiomontanus: His Life and Works” (or, rather, to Ezra Brown’s 1990 translation of the same).

On the one hand, Zinner is crushingly magisterial, in both tone and deep attention to detail (though the appendices by other scholars do sometimes highlight Zinner’s occasional dependence on others’ unreliable translations). Yet on the other hand, it is a style of writing characterized by what is clearly a passionate drive to understand Regiomontanus within his cultural, scientific, and mathematical context.

Say what you like about Zinner, but you could never accuse him of skimming the surface of the subject: as the old joke goes, he’s definitely more bacon than eggs (i.e. where “the chicken is involved but the pig is committed”).

Peuerbach and nocturnals

I’ve written a number of times before about how I strongly suspect that the circular drawing on f57v had a ring of letter groups that was originally made up of 4 x 18 symbols, but where the first two letter shapes of each set of eighteen were subsequently joined together into odd gallows-like characters to turn the sequence into a (far more mysterious) 4 x 17 letter-group sequence. Quite why the author did that is not known, but it would seem to me to have been done to conceal the extrinsic 4 x 18 structure.

Why, you may ask, would a 4 x 18 ring be a giveaway? This, in my opinion, is because 360 degrees / (4 x 18) = 5 degrees, which is the kind of explicit marking you would see on an astrolabe-like instrument of some sort. And because one of the secret astrolabe-like instruments of the mid-fifteenth century was the “nocturnal”, “nocturlabe” or “stardial” (astrolabes themselves were hardly secret by that time), I have long wondered whether what was depicted here was this specific instrument.

And so I was fascinated to read the following brief note in Zinner (pp.26-26), in his discussion of Georg Peuerbach (Regiomontanus’ mentor and teacher):

From 1455 on, there existed “stardials” [nocturnals] to tell time at night by means of the Pole Star and two stars {the “pointers”} in Ursa Major.

The footnote reference Zinner gives for this is: “Ernst von Bassermann-Jordan. Uhren. Berlin, 1922. Figure on page 21.” And so I went off (eventually) to have a look for this specific edition.

It turns out that von Bassermann-Jordan was a well-known clock and watch collector, whose classic book (“Uhren”) on the subject was reprinted many times. I was therefore delighted to find out that I could order a relatively cheap print-on-demand copy of the exact same 1922 version that Zinner referenced, and speedily sent off my money.

I must admit to having been a little bit surprised when a padded envelope appeared covered in Indian stamps (I must admit to having wondered whether it was a coincidence that “Uhren” was an anagram of “Nehru”), but that’s globalization for you.

Even though the print-on-demand book cover was really quite nice, the quality of the scans inside was unfortunately more than a little disappointing in places. But even so, I could now try to find what von Bassermann-Jordan said on page 21 that Zinner remarked upon. Sadly, this was one of the many places where the scans became somewhat unrecognizable by the right-hand edge.

Yet because I was able to use Google to search for “Orientierung der Horizontalsonnenuhren” on the same page, this yielded three hits on archive.org, including the 1922 edition I had just bought. (There was also a 1914 edition and a 1920 edition). Unsurprisingly, the 1922 edition was (without any real doubt) the source of the somewhat mangled scans for the Indian POD company, so I can show you what I was trying to read, direct from the source:

Even so, this meant that I was able to find the same thing in the (much clearer, and significantly more readable) 1920 edition of “Uhren”, so that you can hopefully see the structure of the nocturnal (dated “1456”) much more clearly:

I presume that this is referring to the Bayerisches Nationalmuseum in Munich, which (encouragingly) has a collection of scientific instruments. However, I wasn’t able to find the one depicted in its object database (and I suspect Zinner wasn’t able to find it either), but perhaps one of my German readers will have more luck than me. 🙂

Other nocturnals

A very good source for the history of the nocturnal is Günther Oestmann’s (2001) “On The History of the Nocturnal“. Oestmann notes that the idea that the nocturnal was first invented in China has been comprehensively debunked, and that it is instead a European invention – there are no Arabic nocturnals from the Middle Ages. There was also a predecessor to the nocturnal (a sighting tube and disk, described by Pacificus in the 10th century), but the more compact hand-held nocturnal was clearly a far more usable version of the same thing.

The “V2.0” idea of nocturnals had actually been discussed as early as the 12th century (though few seem to have been actually built). Raymon Lull mentions the nocturnal in his Nova geometria (1299), and the device was made famous by Peter Apian’s 16th century printed book:

But documentation on nocturnals between Lull and the 1456 nocturnal noted by von Bassermann-Jordan and Zinner seems quite thin. Oestmann lists all the 15th century nocturnal manuscripts he is aware of, together with references (not included here) to where they are mentioned in Zinner’s (1925) “Verzeichnis der astronomischen Handschriften des deutschen Kulturgebietes“:

* Wolfenbüttel, HAB: Cod. Guelf. 81.26 Aug. 2°, fol. 144v (Use of the Nocturnal, Latin 1461)
* Göttingen, UB: 2° Philos. 42m, fol. 55r/v .00 (Johann v. Gmunden [?], Construction of the Nocturnal, in a collection of astronomical texts, 15th cent.)
* Würzburg, UB: M. ch. q. 132, fol. 153v-154v, 155v (Construction of the Nocturnal, Latin, in a collection of astronomical texts, late 15th cent.)
* Leipzig, UB: Cod. 1469, fol. 201r-207r (Construction of the Nocturnal, Latin, in a collection of astronomical texis, 14-15th cent.)
* Munich, Bayer. StB: Clm 24105, fol.65v-67r (Use of the Nocturnal, German, in a collection of astronomical texts, 15/16th cent.)
* Munich, Bayer. StB: Clm 214, fol. 167r-185v (Construction of the Nocturnal, Latin, in a collection of astronomical tables, 15th cent.)
* Bern, Burgerbibl.: Cod. 157 fol. 27v-28v (Construction and Use of the Nocturnal, Latin, 15th cent.)
* Zürich, Zentralbibl.: Ms. C 107, fol. 107r/v (Wilhelm Hofer, Carthusian monk and pupil of Georg Peuerbach), Construction and Use of the Nocturnal, Latin, Gaming (Lower Austria) 1472/79); see L. C. Mohlberg, Mittelalterliche Handschriften (= Katalog der Zentralbibliothek Zürich, vol. I), Zürich 1951, p.55f., 361).
* Ottobeuren, Klosterbibl.: Ms. II, 319, fol. 122-123 (Construction of the Nocturnal, 15th cent.)
* Meiningen, Landesbibl.: Pd 32.44, fol. 92r-98v, 104r-105v (Construction of the Nocturnal, 15th cent.)

Oestmann also mentions Chartres MS 214 (olim 173; destroyed in 1944), which looked like this, though note that it actually belongs to the earlier “sighting tube” tradition from Pacificus:

But because Oestmann relies on Zinner, who in turn was only looking at astronomical manuscripts within the German cultural orbit, there are doubtless many more to be found. For example, there’s a nice volvelle nocturnal in fol. 25r of MS Ashmole 370, an English manuscript dated ~1424 that I’d really like to see the rest of one day:

If anyone is aware of a better / more recent / more pan-European source on the 12th-15th century history of the nocturnal / nocturlabe than Oestmann, please let me know!

“Stretched out arms”

For Voynich researchers, I would argue that the single most extraordinarily interesting paragraph in Oestmann’s paper is as follows:

Closely connected with the history of the nocturnal are certain diagrams in nautical texts, which served as mnemonic devices for the correction of the measured altitude of the Pole Star. The position of α and β Ursae minoris respectively α and β Ursae maioris, often called ‘Guards’, indicated the correction to reduce the observed Pole Star altitude to obtain latitude. A man with stretched out arms standing in the Pole was used. If the Guards were found over his head the Pole Star stood 3°5′ under the Celestial Pole and vice versa. The two arms marked the side deviations and also intermediate positions were recorded in mnemonic verses.

Given that f57v specifically depicts people with stretched out arms (at the top and the bottom), I suspect that this is an avenue of research that is well worth pursuing further:

Oestmann’s footnote #17 gives as his source:

See for example the Regimento do Norte (probably composed in the 15th century; München, Bayer, StB: 4° Inc., 1551). On the nautical ‘Regiments’ see Joaquim Bensaude, L’astronomie nautique au Portugal à l’époque des grandes de couvertes (Bern, 1912/17, repr. Amterdam 1967), pp. 136-145, 223f.; Hermann Wagner ‘Die Entwicklung der wissenschaftlichen Nautik im Beginn des Zeitalters der Entdeckungen nach neuern Anschauungen’, Annalen der Hydrographie und Maritimen Meteorologie, 46 (1918), pp. 215-220.