I’ve recently had a number of emails from Don of Tallahassee, describing various ways in which he thinks Voynichese can be decomposed into simpler subunits: broadly speaking, his scheme is similar to Jorge Stolfi’s well-known crust-mantle-core model, but with a very much larger base group.

Numerically, Don’s model works well: but – in my opinion – it doesn’t yet help us move towards what I would consider any of the basic milestones we would need to pass before we can crack the puzzle of Voynichese.

If anyone wants to be the Voynich Champollion, here is my list of the milestones you’ll need to tackle in your research programme, with various sample challenges. I don’t mind admitting that I haven’t yet succeeded at any of these: make no mistake, they are all hugely difficult.

(In the context of Don’s models, my opinion is that he – like many others before him, so it is in no way a criticism – has effectively skipped over the first three milestones, and gone straight for the modelling milestone. But we all need to get vastly more confident about the first three milestones before we can start doing modelling in an effective way.)

Milestone #1: Reading

Personally, I’m not convinced that we’re even reading Voynichese accurately off the page yet.

For example:
* Page-initial letters have quite a different instance frequency distribution from anything else, particularly in the Herbal pages. Why should that be?
* Line-initial letters have, again, a different instance frequency distribution as compared to text within lines. Is it therefore safe to assume that these are the same kind of text as each other?
* In 2006, I proposed that EVA ‘aiin’ characters may well represent Arabic digits, by steganographically enciphering the values using different shapes of the scribal flourish on the tail of the (‘v’-shaped) EVA ‘n’. This basic hypothesis needs to be tested microscopically and with careful imaging techniques, but my proposals to the Beinecke some years ago to do this were turned down.
* Philip Neal has pointed to evidence that certain stylized text sequences may be quite different from the rest of the text. There are both ‘vertical Neal keys’ (down the start column of many pages) and ‘horizontal Neal keys’, which often appear about 2/3rds of the way across the top line of a page or paragraph, and often ‘bracketed’ by a pair of single-leg gallows (‘p’ or ‘f’).
* In 2006, I proposed that Neal keys might form part of a tricky in-page transposition cipher (as described briefly by Alberti in 1467), where the gallows characters might form references to within key-like sequences. But this hypothesis has not been tested any further.

Challenge: when we try to decipher Voynichese, are we even trying to decipher the right thing? When there are so many different things that each suggest that the text as a whole is not an homogenous entity, why do so many people persist in treating it as if it is a single, simple language?

Milestone #2: Parsing

The second roadblock is that we can’t yet even parse Voynichese. Because of the ambiguities and weird letters, Voynich Manuscript researchers use a stroke-based transcription called ‘EVA’: this lets us transcribe the text and talk about it, even if we disagree (or are uncertain) about how these should be parsed.

For example:
* Is ‘ch’ a unique letter or is it a ‘c’ letter followed by an ‘h’ letter?
* Is ‘ii’ a pair of ‘i’ characters or a separate character?
* Is ‘ee’ a pair of ‘e’ characters or a separate character?
* Are ‘cth’ / ‘ckh’ / ‘cfh’ / ‘cph’ actually a t/k/f/p gallows character followed or preceded by ‘ch’, or four entirely separate composite letters?

Challenge: what kind of statistical tests would help us compare multiple different candidate parsing schemata, to help us decide which ones are more likely?

Milestone #3: Tokenization

The third roadblock is that there seems strong visual evidence that characters are not the same as tokens: which is to say that some individual letters in the plaintext may map to multiple letters in the Voynichese ciphertext.

For example:
* Is ‘qo’ a token?
* Is ‘dy’ a token?
* Is ‘o’ + gallows a different kind of token to just plain gallows?
* Is ‘y’ + gallows a different kind of token to just plain gallows?

Challenge: what kind of statistical tests would help us compare multiple different candidate tokenization schemata, to help us decide which ones are most likely?

(Note that Milestones #2 and #3 overlap sharply, making the process of getting past them quadratically more difficult, in my opinion.)

Milestone #4: Modelling

Even if we get to the stage that we are able to read, parse and tokenize Voynichese with some degree of certainty, we still face many grave difficulties, not least of which is that we have at least two ‘dialects’ to solve at the same time – Currier A, Currier B, and ‘Labelese’ (for want of a better term). For each of these languages/dialects, we need to model the language functioning and use the results to understand their internal structures.

For example:
* What do the contact tables between adjacent tokens suggest?
* Can we produce Markov models for each of these ‘languages’?
* Is ‘qo’ a free-standing unit (i.e. that is only steganographically prefixed to words), or is it genuinely an integrated part of words?

Challenge: what is the mapping between Currier A, Currier B, and Labelese? Can we somehow normalize the three such that they all conform to a single unified scheme? Or are there basic differences between them such that this is impossible?

145 thoughts on “Who will be the Champollion of Voynichese?

  1. Although I do believe it is helpful to group such problems thematically, I’m not sure that the milestones need to be strictly following. Research on the Voynich manuscript is a feedback process and sometimes getting ahead of ourselves throws useful light on the things we do not yet understand.

    If we think of Egyptian hieroglyphics, for example, identifying the role of cartouches came way before any understanding of the text or script. Likewise, the earliest attempts to read names in cartouches came before understanding how characters were spatially related. It was only through attempting to fit the script into a crib that character order could be established.

    (Similarly, Champollion put his money on Coptic being a related language before there was much proof of that. I wouldn’t want to approach the Voynich in that way, however.)

    So long as a problem can be typified it can be contained. We know that Grove Words are different from the text as a whole and thus should be wary of using them uncritically. That a recent decipherer has failed so majestically to beware Grove Words might give us a good laugh, but should not give us pause. Knowing why Grove Words are different would give us a useful insight, but knowing that they are and how they are is enough (or at least enough to stop up from believing that the herbal section is about cotton, crocus, cardamon, centaurea, etc, etc).

  2. Anton Alipov on September 13, 2015 at 1:41 pm said:

    All good points.

    For point #2, the recent article by Brian Cham is generally on the right way, I think (although I expressed a lot of low-level critique to that).

    The character level is not the optimal starting point anyway. It’s better to start (if we are still starting :)) with attempts of tracing the narration structure and detecting contextual dependencies. If any positive results are achieved, this will then allow us to step down to the character level. I recently applied this approach to investigation of “Voynich stars” (be they stars or otherwise).

  3. Emma May Smith: I listed the milestones in such a structured way because it seemed to me that the overwhelming majority of people who have landed on Planet Voynich in recent years seems not to appreciate how problematic Voynichese is, and also on how many different levels those problems are scattered. (If I had included Grove words and other similar problematic features, the post would have doubled in size!)

    Personally, I am sure that finding workable cribs will drive at least some of Voynichese’s eventual decryption: but I also think that all the time we can’t read the strokes, parse the letters, group the semantic tokens, or even partially model the semantic structure of those tokens, that blessèd day will necessarily remain a very long way off. 🙁

  4. Anton: can you give a link? I don’t recall seeing anything from BC recently. =:-o

    I would certainly agree that we have to do top-down stuff, such as my “block paradigm” matching. But to stand any real chance, I think we have to understand Voynichese itself (i.e. in a bottom-up way) at the same time.

  5. Hello Nick,

    I couldn’t agree more, in particular with points 2 and 3.
    They are excellent criteria to judge (hmmm… reject) proposed solutions (even if one usually can’t convince the proposer of this…..).

    I would add the point of Eva-f and Eva-p. These could represent several different things (alternate forms, ornamental nulls), but they *don’t* represent standard letters in the Voynich (or any other) alphabet.

  6. Anton, I think Brian Cham picks up on some interesting aspects of character formation, but I cannot buy his theory. Both EVA: g, b are rather rare, while the curve characters EVA: d, ch, sh have no real counterparts (unless we count even rarer characters).

    But I think my main concern is simply that proposing EVA y, l are counterparts runs into the problem of why their distribution is so different. Though “al, ol” are very common, “ay, oy” are not. He quite rightly instead treats EVA: l, r as counterparts in considering nonconformity within his system.

    However, I do agree that curve and line are valid groupings in some respects as they condition the appearance of EVA: a, y, which I consider variants. But this runs against his system so I do not know how they could be brought together.

  7. Anton Alipov on September 13, 2015 at 9:23 pm said:


    As I noted, I myself expressed to Brian virtually *a lot* of critique, which is only too long to be posted in a comment. At present, his developments are in rather undeveloped state indeed. But there is rational and promising approach there, and I believe it can be made truly systematic.

  8. Rene: given that the issue of EVA f and p falls squarely into Milestone #1, perhaps you’re not so far away on all three milestones, not just #2 and #3. 😉

    More generally, do you know if anyone has written a paper specifically on f and p?

  9. Nick,

    the first name that comes to mind is Philip Neal. I couldn’t point to any specific document (online or otherwise) though.

    I’m also not sure if he was the first to note the odd behaviour of Eva-f and Eva-p (also in their normal forms), that they almost exclusively appear on first lines of paragraphs.
    Almost, but not 100%…

  10. Thanks, Nick. well said.

  11. Milestone #1. Fragmentation reduces the quality of stats. Lumping lessens the value of stats. Neither completely destroys all information gained by counting and analysing. In some instances it is immaterial (but I won’t give examples). Choices are to accept the conditional nature of some results (not all) or give up.

    Milestones #2 and #3. Don’t decide how glyphs were intended to be parsed. Try possibilities separately. If one or more possibilities result in resemblances to a language, you can chase that shadow. If you find unambiguous and truly sensible paragraphs in, for instance, ancient phonetic jadhistani, you have reached the top of the glass hill. Consistent string transposition and who-knows-what will be necessary. It’s a great challenge.

    Milestone #4. We should have taken Emma May Smith’s advice. Also, keeping in mind: The location of gallows glyphs with much variation, the selection of line initial and line terminal glyphs, the page format made to resemble pages of a “real” book; all appear to be art and contrivance. Glyph harmony has long been recognized and can be described without an essay, in one short paragraph. If you look for steganography or cipher in all that, you are just having fun. Until some kill-joy discovers the author’s notes, why not?

  12. D.N.O'Donovan on September 14, 2015 at 5:06 pm said:

    qwerty has 9:26.
    EVA has ?:37

    does this matter?

  13. Out*of*the*Blue on September 14, 2015 at 8:54 pm said:

    These technical methods of analysis are all well and good, something that needs to be done. But if we presume that the language of the text is intended to convey meaning, then there needs to be a way to discover the cryptic composition method used by the author. Should we presume that the author intended the text in this way. Can it be read at all?

    Don of Tallahassee’s recent proposals are an interesting combination of cataloging with the three part theory of Voyhichese ‘words”. Since the text contains such a number of plants identified by some researchers, it is perhaps not unreasonable that ‘words’ turn into pharmaceutical formulations. I have a theory that if the pages showed illustrations of automobile parts, the same cataloging system could be used to create a universal directory for car parts without any changes to the original VMs text.

    Does the author intend for the text to be read? If there is meaning to be found, it is well covered over. The complexity of the composition (reading) method is irrelevant if the secret of the process is obviously revealed in some other part of the document. However there is a secondary part to the translation besides the matter of HOW. Along with the general matter of method, there is also value in the specific text itself, the questions of What text and Where is it found?

    Interesting and unique unto itself is the number of triple and paired ‘words’ in the outer text band of VMs (f71r) White Aries, when compared with all other such bands in the VMs Zodiac. Stranger still that this is the text passage that is designated as the papal golden key by the pictorial investigation of the astrological and heraldic aspects of the pairing paradigm as illustrated in the VMs Zodiac. These two independent factors both designate the same text segment.

    Method is not the only determinant of success in mining for gold. Success also depends on location. Perhaps similar considerations could be applicable to the VMs.

  14. OOTB: jumping forward to pharmaceutical formulations – or indeed to any other model – without first understanding the fundamental aspects of the is nothing more than an exercise in prematurity. Similarly, feel free to wonder to yourself whether or not the text can be read: personally, I’d rather be working with it than speculating about it.

  15. Rene: in almost all ways I can think, the artificial behaviour of EVA p and f offer an immediate practical counterexample to the notion that Voynichese is a ‘raw’ language. For if p and f are artificial constructs, how can Voynichese possibly be a raw language?

  16. Emma May Smith: I’ve had a good read of Brian Cham’s paper, and I think it is (unfortunately) a prime example of how badly people can get led astray by relying too heavily on a stroke transcription.

    The cornerstone his paper is supported by is the apparent stroke-to-stroke ‘harmony’ offered by the aiin and aiir families: but if these are actually single cohesive tokens (i.e. ‘aiiv’ is a token, etc), that whole harmony vanishes. In many other places (such as ar / or / al / ol), that claimed harmony is very much more debatable: and ‘q’ doesn’t fit anything. Sorry, but that’s how it looks to me. 🙁

  17. Nick, I won’t lie, as somebody who believes in a linguistic solution to the Voynich manuscript EVA f, p give me the willies.

    Not so much Grove Words, which can be explained away easily, but rather Neal Lines. I do not have a good explanation for them.

    I don’t think they’re catastrophic to linguistic theories but they are something that need to be acknowledged. Theories which don’t address them won’t, in the long run, pass muster.

  18. I once wrote a short paper showing statistical evidence that EVA f and p are both allographs of k (not of t) and are predominantly found where the first line of a paragraph has blank space above it. I know that I showed it to Nick but I can’t remember if I put it on line anywhere.

    For the past few years my hobby energies have been devoted to another project which is now complete and I mean to come back to the statiistics of Voynichese in the near future.

    A point I think I have made before is that the Voynich cipher (if such it is) is verbose, so that a paragraph of, say, 5 lines in the manuscript would very likely only have occupied 3 or so in the original. If so we might be able to detect traces of the fact in the structure of the ciphertext.

  19. milestone #5: ACCEPTATION of the proposed solution

    The acceptation, of the proposed translated text, by the “main amateur Voynich research community” and by “professional transcribers” is one of the biggest challenges not mentioned.

    Nobody cares and nobody knows that many amateurs are researching the MS possibilities and could come to a point that a “fair” transcription & translation can be given.


    what are the criteria for the acceptation of that text?

    Several months ago, i managed to translate some arbitrary lines entirely, but it is such a free translation, that i could either translate it to a herbal text or translate the same lines into a religious text. Then i stopped because it did not feel “logical”. I could continue to do that with many lines, but then i faced big problems at the words in the pictures: they then can not be translated into nouns or names, but are mostly nonsense or verbs. Also that did not feel “logical” to me.
    Now, i am working on a small generic internet tool where anyone can see the MS (manuscript) and the text proposal and if one whishes, make their own transcription and translation. Such a tool did not exist for MS in general other then collaborational tools for scholars where the focus lies on the transcription and not the translation.

    I did not read everything above, because i do not want to go into detail discussions with other people of which i do not know how well-educated in this matter they are, but would really like a good discussion on details of the MS somewhere. There is a big difference in people that are only posting text-messages, and people that are really spending time in research and analysis.

    I think a forum-like would be better then here as a reply message in a Blog. For example the “EVA f and p” is one subject i investigated thoroughly and have graphical and statistical information to show the behaviour of those letters and some proposed solutions. Such a substantive discussion can not take place here.
    Such a forum would help enormously, because we can then slowly build towards an accepted solution, together. Now, all what is left is some small messages, some tweets and sometimes a fundamental thingie that is placed between other messages…

    I can host, set up, and pay for such a proposed forum, but i will Not maintain it nor administer it. If somebody thinks this is a good idea and seriously wants to take care of that, please drop Nick a message he can forward to me.

  20. David: I think a good starting point for accepantance of any “Milestone #5” would be having some kind of rationale for each of milestones #1 to #4. As yet, people seem eager to claim the finishing line without doing the hard work of actually running the race. 😐

  21. Philip: I think you showed it to me rather than sent it through, I don’t seem to have a copy of it here. 🙁

    The subtle differences between t, k, p, and f are something people haven’t (as I recall) investigated in great depth. As I (kind of) alluded to in the post, statistical evidence hasn’t yet really been the huge help one might (quite reasonably) think it would be: and I’d be happy to support anything that might turn that round. 🙂

    On the issue of EVA f and p, I once wondered whether the space at the top of a paragraph might instead be an artefact indicating that these horizontal key-like sequences were actually written first, with the remainder of the line then added afterwards – i.e. what you’re suggesting but in a reverse order. I’m not sure I ever looked at this any further, though: something to consider, anyway. 🙂

  22. Anton Alipov on September 16, 2015 at 9:11 pm said:

    There is the master’s thesis of J.B. Morningstar “Gallows Variants as Null Characters in the Voynich Manuscript”.

    Also I remember reading some article stating that “p” is in a certain way statistically different from other gallows, but I can’t find it again…

  23. Anton: Jason Morningstar’s thesis was, in my opinion, a bit too narrowly focused. I would have been more impressed if he had built more general machinery to analyze arbitrary ciphertexts for the presence of nulls, rather than specifically concentrating on the gallows. As we saw (not that many years ago) with the Copiale Cipher, there is still (I believe) no automated way of assessing whether individual characters are nulls.

  24. Anton Alipov on September 16, 2015 at 10:27 pm said:

    Well, a master’s thesis is a master’s thesis, no more, no less.

    As one professor used to say when I was a student, one should not be criticized for what he has *not* done in his paper, only for what he has. 🙂

  25. Dear Mr. Pelling,

    When I first started investigating the Voynich Manuscript, I went looking for plant names in the text.

    Since I was pretty sure what the the plant on folio 16r was, I went looking for the name hashish, the one name I was sure had come down from before the dating of the VMS – Marco Polo had mentioned the Old Man of the Mountain and his fanatic followers, the Ahashishieen (the Assassins), as had many accounts of the same story from Crusaders returning from the Holy Land. The Ahashishieen were said to be made ready for their duties and possible self-sacrifice by a drug, hashish, and a garden of women and wonderful things. Ahashishieen means hashish eaters/users.

    I went looking for the patterns of glyphs/letters/sounds of h-a-s-h-i-s-h and h-a-s-h-e-e-s-h (the patterns A-B-C-A-D-C-A and A-B-C-A-D-D-C-A).

    No luck with them, but…

    Lo and behold, in the first glyph word on the second line was the pattern A-B-A-C-B-A (something like h’shish, which, it turns out, is an acceptable pronunciation/written form.

    Not convinced, huh?

    The last six glyphs, third line from the end also form the same pattern A-B-A-C-B-A.

    It isn’t a common pattern in the VMS.

    I look at coincidences with a judicious eye.

    Still not convinced?

    On the next line (second from bottom), if a downstroke is added to the first o glyph (to form a g glyph), the second through seventh glyphs also read the same pattern A-B-A-C-B-A. (I think this missing downstroke is a bit of intentional deception.)

    In fact, if a reader starts at the end of the third to the last line, I think that, using my proposed phonetically pronounced glyph/sound alphabet, the words should read, h’shish ah’shish i(ng)h inih ir nsreha preser(u/v) at ah ilh. Or hashish, ahashishieen in iron (ihirn) Syria preserve at a hill.

    That’s nine words in a row to form, what seems to me, a relevant and somewhat intelligible sentence about the fortress (iron Syria preserve) of the Old Man of the Mountain.

    Yeah, it’s a stretch. But remember, this is just a small adjunct (and the base) to the rest of my theory, which seems to show most VMS words are entirely different – these readable ones only put in to offer a key to decryption of the manuscript (or maybe to give a few readable sentences for inspectors/inquisitors/snoopy eyes that didn’t give the rest of the secret away, I dunno’).

    This is as many words as some of the real authorities claim to have deciphered, I think, but these are all from one place in the ms.

    My proposed solution gives individual sounds for each glyph, always the same, as shown in Table VII at my fumbltdiddles.com site at:


    If still not convinced, ask anyone who has ever grown a pot plant what the image on folio 16r is.
    I got 10 answers, all the same, no hesitation. (I was surprised by a couple of those that admitted they had had a wilder side in their younger days – they appear much too staid, gentile and refined now.)

    I have built my whole ungainly, asymmetrical, shaky-as-an-Autumn-leaf, house-of-cards proposed solution starting on the sounds incorporated in the glyph word pronounced h’shish.

    Starting with sounds also means I came the long way around to find my proposed solution.

    Thank you.

    Don of Tallahassee

  26. Dear Mr. Pelling,

    Thank you for mentioning my work in your blog.

    I have never heard of Grove words before tonight, so please forgive me if I misunderstand something.

    As always, the following is contingent on my proposed ideas being accepted by others.

    The VMS glyph words seem to be arranged on the pages (generally) in several, or more, individual recipes, sometimes multiple recipes per paragraph.

    The predominant Grove word-starting glyph is EVA = p, if I have this right. In my proposed solution, almost all words starting with that glyph are individual ingredient measurements of one of the following herbs:

    zingiber (ginger), Zingiber officinale
    zacinth (jacinth, blue bell), Scilla nutans
    zizany (corn cockle), Agrostemmna githago
    zitterwurzel (zedoary, turmeric), Curcuma zedoaria
    zirwan, Lolium tementulem
    zucker (zuccara, sugar cane), Saccharum officinarum
    zafara (saffron), Crocus sativa
    zirbel (stone pine), Pinus pinea
    zerumbet, Curcuma zerumbet
    ziziphus (jujube), Zizyphus vulgaris
    ziziphus lotos, Zizyphus sativa

    The codes for zingiber, zucker, zafara, ziziphus and ziziphus lotos seem to appear most often in the VMS words. I don’t think the others show up very often.

    It seems reasonable these herbs might show up most often because they were probably used in more herbal recipes.

    I think the recipes are arranged on each herbal page with the recipes with most expensive ingredients (for the royals/nobility) at the top, then the less expensive recipes for the richer citizens and so on down to the recipes consisting of cheap ingredients (for the poor people). The ingredients of zingiber, zucker, zafara, ziziphus and ziziphus lotos should mainly appear in the topmost or upper recipes because they were imported (I think) into England and were probably expensive.

    It wouldn’t seem too much of a stretch to think ingredients inside the recipes might be listed in any order the author deemed most acceptable to him/her – maybe an order that tried to place a Grove word at each page’s start, maybe paragraphs as well.

    I think the dividing point between contiguous recipes is sometimes where a word is repeated – because why would an ingredient be repeated in a recipe? It also makes an easy way to not have punctuation dividing the multiple recipes in some paragraphs. (The other way to distinguish where recipes divide from each other is when an herb already shown earlier in a recipe is shown a second time – dividing line before second occurrence.)

    This is my attempt at an explanation that seems to reflect what the VMS words show with the Grove words.

    I think the beginning-of-line-glyph statistics are also somewhat related to the author’s-choice of order of words, as are the appearances of many EVA = m or g ending words at lines’ ends. In most recipes, it wouldn’t matter which ingredient was listed or added in what order. But it might help confuse outsiders if done in unexpected ways.

    Not having a Grove word available in the first recipe on a page may explain some of the really strange complex glyphs, those stretching over two words.

    One thing I noticed – many page-initial words seem to be two words separated by a small space – only found normally in some page-initial words. I think this might be why the statistics are strange.

    I would appreciate any corrections to my understanding or delusions of reality.

    Thank you.

    Don of Tallahassee

  27. Don: linguistic decryption is a difficult road to pursue to the end, with or without the aid of hashish. 🙂 For me, Bax’s attempt at decryption does little but recapitulate numerous badly broken attempts that came before before him, so I wouldn’t be comfortable classifying that as “authoritative”, nor using it as any kind of yardstick. 🙁

    I’ve written elsewhere than just about any random way of approaching the text will probably yield roughly ten matches – for example, there are roughly ten English words in the EVA transcription, but it would be a mad brave man who would use that observation to prove that Voynichese is actually English.

    To my mind, you’d need to do a lot better than what is essentially a null hypothesis to be able to convince other people (not just me) that you’ve even come close to cracking it. Otherwise you’ll end up like Bax, a forlorn figure unable to pull back the unwise tide of Voynich translation press releases he pushed out into the world.

  28. Out*of*the*Blue on September 17, 2015 at 8:44 pm said:

    Hi Nick,

    You said: “OOTB: jumping forward to pharmaceutical formulations – or indeed to any other model – without first understanding the fundamental aspects of the is nothing more than an exercise in prematurity.”

    Fundamental aspects of the xxxxx? Language -or similar, I presume. In my opinion, Don’s fundamental problem is the failure to make an absolute, ideological connection with the text. His presumption of content, which then drives the cataloging process, is only a reasonable conjecture.

    You also said: “Similarly, feel free to wonder to yourself whether or not the text can be read: personally, I’d rather be working with it than speculating about it.”

    Implying that I shouldn’t wonder here?? But it was not the question of whether the text CAN be read, rather a question of whether the text was INTENDED to be read by other individuals, potentially those who were not (initially) privy to the secrets of the VMs linguistic system. I didn’t make that part very clear. And if intended to be read, then how? Would you contend that the author intended for the VMs text to be read by the system you have proposed? Would the author expect and intend the text to only be readable by computerized statistical analysis? If not, then there must be some other method, which may have been hidden in the ms.

    I’m not saying such analyses are not valuable, but it is no secret that they have yet to solve the problem. So looking at the other side of the coin, is there a potential method of understanding that follows a path which conforms to what the author intended? A clear series of clues exists in Zodiac illustrations.

    It seems to me that the analytical criteria applied to the statistical summation of the text as a whole could be entirely different, perhaps even irrelevant to what might need to be applied to a specific segment of text taken from the VMs. Or maybe not?

    I’ve already examined the primary segment in question, the outer band of f71r, and found it to contain several word repetitions, with the presence of such repetitions as an almost absent factor in all other VMs Zodiac text band segments. What other methods of comparison and examination should be applied?

    So as to your speculating about the text, how that is done will certainly be important, but as I see it, what is speculated about, certain, specific, text segments identified by the author’s visual narrative within the VMs itself, may also be critically significant. Yet, somehow, IMHO, the whole concept escapes consideration.

  29. Re. nickpelling September 16, 2015 10:02 pm:
    “… there is still (I believe) no automated way of assessing whether individual characters are nulls.”

    Brian Tawney wrote a program that “scores each byte in the text on the basis of how much the frequency of its neighbors
    conforms to the average throughout the text. The higher the score, the more likely the character is a single uniliteral null. This would not identify multi-character nulls, or nulls which are only null under certain circumstances.”

    If I remember correctly, no EVA letter in the VMs was flagged as null. EVA-x might have been suspicious. And the untranscribed glyphs, naturally. We can call nulls that are consistently placed adjacent to certain letters to avoid detection “intelligent nulls”. For relative scores, Herbal B in the first quires is much nearer Herbal A than it is to Quires 13 and 20, FWIW.

    Incidentally, by this measurement, all characters in Z340 are “definitely not null”.

  30. Knox: bear in mind that nulls also depend on the transcription & parsing model you assume, so you can’t (I think) trivially plug in EVA and expect it to work.

  31. OOTB: wondering about whether or not the author intended the text to be read doesn’t seem to me to gain much. In general, I find speculations as to authorial intent more of a hindrance than a help – rather, the trick is to understand the text more on its own terms and work backwards from there.

  32. SirHubert on September 18, 2015 at 9:10 am said:


    Personally, my problem with discussion of a ‘visual narrative’ throughout the manuscript is that it implies a single original author with a single unified purpose. I can’t see any particular reason to assume any of these things.

    In fact, we know that at least two (possibly more) scribes physically wrote it (Currier and others) and that there appear to be two ‘languages’ or encryption/generation techniques used (also Currier).

    If what we have is a compilation of material, some perhaps original, some taken from a variety sources which may be chronologically and geographically disparate, you clearly cannot expect to find a coherent ‘visual narrative’ throughout. In individual sections, possibly. Throughout, no.

    I’m showing my age, but about thirty years ago the Goodies did a superb spoof of Erich von Daniken in which they demonstrated, among other things, that the sombrero and the lemon squeezer were both copies of flying saucers, that a bishop’s vestments took their form from a multi-stage rocket.

    The problem with the Voynich Manuscript is that there is so little to act as a control that you can argue fairly plausibly for all sorts of things. Not that this is confined to the illustrations – the text suffers equally, alas.

  33. SirHubert: The Goodies, blimey. Here they come, into town, getting up, falling down.

    There’s a lesson in there for all of us. 🙂

  34. Nick: As I said, stats are conditional on input. I doubt there will be much difference in the results of the null program if I use any of the several dozen variations of basic EVA in my files. However, if we run the program on Voy-101 and extended EVA they might. The additional glyphs in those transliterations that do not score differently from the ones they most resemble in basic EVA likely are scribal variations.

  35. Out*of*the*Blue on September 18, 2015 at 7:32 pm said:

    Nick & SirHubert,

    “Wondering”, and I believe you introduced the term, just wondering is not going to be of much help unless it opens other perspectives. My question was to the author’s intention. Was the text intended to be read? Yes or no? If no, then pack it up and put it away; go out and mow the lawn or solve some of the world’s other pressing problems? If yes, then HOW would that be? Does the author provide a way by which the text can be read? And if so, can it still be found and used as intended. I had thought the analogy would have been a bit more obvious. What would Champollion have accomplished without the Rosetta stone? Method is great. Method is essential. But method has to be applied to something. And if there is not something, the right something, for the right method to be applied to, then there may be no solution. That seems to be the current situation. Maybe, however, the solution is not to be found by applying ever more complex and anachronistic methods to the statistical conglomerate of VMs text. Maybe the solution could be found by applying appropriate methods to a particular segment of text. What segment of text might that be?? One that the author has provided as a result of following the VMs visual narrative, an internal, text delivery system.

    I said nothing about a visual narrative extending ‘throughout the manuscript”. Instead, I did specifically mention the VMs Zodiac. And the way to tell that it is an intentional narrative is that it is something too complex to occur by accident and that it is historically grounded in the real world.

    Look at the VMs Zodiac:
    Pisces first, contradicts traditional arrangement, Pisces is an easily overlooked PAIR.
    Aries is split in two. Outrageous!
    By the bye, two halves make a PAIR.
    Taurus is split in two, More outrageous, Another PAIR.
    Gemini is a natural PAIR.
    Cancer is an uncommon PAIR.

    Aries and Taurus are split PAIRS. Therefore a PAIR of PAIRS.
    Pisces and Cancer are PAIRS of aquatic animals. Also a PAIR of PAIRS. And a second PAIR of PAIRS, which makes a PAIR of PAIRED PAIRS. And a PAIR, which is Gemini.

    Does this look like an accidental concatenation of circumstance to you? And as the manuscript is an artifact of human creation, if not accidental, then what????? But wait, there’s more.

    The visual narrative then establishes an ideological narrative based of pairing; the pairing paradigm. And derived from a passage in Deuteronomy, that facts must be established on the basis of testimony from two or three witnesses. Minimum number of witnesses: two = a PAIR.

    Where does the pairing paradigm go next? Look at the top of the outer ring of figures at the patterns of the tubs in Pisces. There is a PAIR of matching patterns, flanked by another pair of matching patterns. A PAIR of PAIRS right off the top. Consider looking and basic heraldic patterns – not too obscure for the Middle Ages, you might think. A PAIR of papelonny patterns is split between Pisces and Dark Aries. A PAIR of blue-striped insignia is found on White Aries. Try seeing these in the absence of any radial influences. And (voila!) you have a potential PAIR of armorial insignia historically representative of the Fieschi popes of Genoa. (With a little help.) Look more closely and you’ll see the red galero of the inner figure which illustrates the historical event when Pope Innocent IV made his nephew a cardinal (1251 CE). And he was the second pope. Innocent IV also began this tradition of the red galero in ecclesiastical heraldry. Note the proper hierarchical placement of the two figures in the celestial spheres. Note the position of both figures in the favored upper right heraldic quadrant. (And there’s more.) Note that positional elements are objective, not subjective. That is, if and only if the subjective identification is valid, do the objective positional confirmations make sense.

    The blue-striped pattern of the outer figure is directly connected to another pattern sitting in the the circular band of text. This is a unique construction – only in the f71r illustration. A second pattern is found in the outer band of text just above this figure’s hand. The cardinal has the red galero. The pope has a PAIR of papal keys, one golden, one silver – a key to heaven and a key to earth. These patterns easily serve as textual markers and the outer segment has multiple examples of paired and triple repetitions of VMs ‘words’. What can this mean? Seriously.

    The thing about the pairing paradigm, I believe, and I may be sticking my neck out here, is that it seems to be infallible. The correct reading of the visual narrative in the VMs Zodiac must be marked by pairs. The intended pathway is marked by pairs. So the investigation must follow the paired pathway. Either stay between the pair of lights or it’s not going to land on the runway. The author has intentionally reconstructed the traditional medieval Zodiac sequence to set up the pairing sequence needed to establish the paradigm. Just like everything on White Aries is painted in order to show that the animal is white.

    And did I mention the papelonny pun? It also occurs as a pairing and as an objective positional confirmation. That is, the papelonny patterns of Pisces and Dark Aries match in sphere and quadrant the positions of the blue-striped Fieschi insignia that are found on white Aries. And the clincher is that ‘pape’ is French for ‘pope’. Pairing is demonstrated in the identifications on White Aries and again with the papelonny mouse trap. Now the identifications are also labelled and PAIRED.

    The identities are secured by the number and complexity of the confirmations, but at the same time it is only obvious that the VMs representations, hidden by radial influences etc., were definitely not intended to be the clearest, most flattering portraits. The obfuscation is mostly intentional, two out of three. There is more here than meets the eye. And that can clearly be shown by the example of papelonny that the author is far more sophisticated in the use of heraldic images than many prior and present researchers, but on the whole it is a text delivery system, a recoverable text delivery system which the author has placed in the text. Now, what to do with the text that is delivered? What might actually work?

  36. OOTB: good luck with finding someone willing to read that kind of comment from start to finish.

  37. Out*of*the*Blue on September 18, 2015 at 10:35 pm said:


    I see obtuseness and mental blindness have a number of similarities. Every point I have cited has a standard cultural (heraldric etc.) or historical basis in fact. Or else it is simply and plainly visible in the VMs. But apparently you won’t be bothered. How long have all VMs investigations passed over heraldry in general and papelonny in particular without any realization of what the text clearly contains?

    I could offer you a structural comparison of our respective heraldic identifications, if you prefer that, but actually explaining the details would certainly prove too lengthy.

    A serious weakness in your hypothesis is that you seem to be establishing an identify only by using the potential charges in the plant roots, which happen to be the two examples most commonly used in heraldry (eagle & lion). To my knowledge you’ve made no consideration of tincture, either of the charge or of the field. And any change in tincture causes a change in identity. And if not specified, it could be anything. At least in White Aries the stripes are clearly blue. Apparently, in your interpretation, any other colour would serve as well. And that is simply where you are wrong.

  38. OOTB: your cryptoheraldic interpretation of one small section of the Voynich Manuscript remains speculative for a whole variety of reasons, some of which you have acknowledged, many of which you have not. Over the last century, countless interpretations of other small sections of the Voynich have been put forward which are equally speculative: none of them has, so far, been of any use in decrypting the text.

    There continues to be a wide gap between Voynich speculation and Voynich decryption: simply repeating your papelonny speculation in comment after comment here does not narrow this gulf.

  39. Out*of*the*Blue on September 19, 2015 at 8:03 pm said:

    With all due respect and a sincere appreciation for your providing this opportunity for discussion, I must say that I still have significant disagreements with your replies. When have you asked a question or made a comment that is relevant to the heraldic or historical topics that are reflective of those contained in the VMs visual narrative that I so drolly recounted just the other posting back? Those are the details that are found in the VMs Zodiac illustrations. Sorry it is sooo long and boring. It is the clear existence and complex interaction of all those details that is compelling evidence of intentional construction of the visual narrative.

    To dismiss all this as a “cryptoheraldic interpretation of one small section of the Voynich Manuscript”, it seems to me that you’ve missed the whole part about the existence of the pairing paradigm and its expression in the first five houses of the VMs Zodiac. Look closely, there is no heraldry here. There is just astrology – VMs weirdo astrology – call it cryptoastrology, if you like, but there it is.

    I get the feeling, as I know you surely have many other demands on your time, that you have never really explored the deeper reaches of the various interpretations of these Zodiac illustrations.

    You apparently also have a problem with speculation. You have certainly misapplied my interpretation of how speculation plays a role in the visual narrative of the VMs Zodiac. Speculation is appropriate in the discovery phase of this investigation. However, when point after point of corroborating evidence is discovered within the illustration, speculation is progressively diminished, until – emphatically- it is eliminated. It’s not just about the identification of heraldic armorial insignia – paired armorial insignia. There is ecclesiastical heraldry, history and so much more – all those details, more than I have cited. The whole complex construction, with historical references, all tucked into one little section. Instead of spread out all over. It emphasizes the interrelationship and the complexity of the elements within the Zodiac illustrations. Nevertheless, (author’s choice) it is the secondary orientation in the White Aries illustration that leads to a significant historical reality. This is not speculation. Does the author’s use of obfuscation appear speculative to you? Does it count against the validity of my identifications? Does speculation on the part of the investigator really imply that the examples are speculative in themselves or rather that the investigator currently lacks a sufficient working knowledge of the whole picture? And then more evidence, previously unknown to all investigations, is found. And what? — Nothing! I fail to understand the disconnection. What examples of mine would you consider speculative?

    As to bridging the gulf, I think we can agree, what we got ain’t working. My investigation continues and, as always, relevant comments are greatly appreciated. My opinion remains that the marked text passages of the VMs Zodiac, as designated by the complex construction I have indicated, were intended to be a significant part of the VMs solution. And not simply on the basis of my idle speculation, but because the author has presented a clear illustration of both relevant patterns uniquely connected together as another example in the list of objective, positional details.

  40. OOTB: I would consider every part of it speculative, and moreover speculation formed from a tiny subset that appears to cast no light upon the very sizeable remainder. I really don’t mind what you do: but please don’t kid yourself that speculation is likely to hand us any kind of magical key to unlocking the Voynich Manuscript. It’s just not that kind of a thing: and the more you try to convince others that it is, the more of your time and their time you’re wasting.

  41. Out*of*the*Blue on September 19, 2015 at 11:19 pm said:

    Seriously, Nick, it’s all speculative? You would consider the presence of blue stripes to be speculative, the presence of two examples of papelonny to be speculative, the proper hierarchical placement of independently identified figures to be speculative, when anyone can easily see these elements in the illustrations. How can that be?

    Or is it just my interpretation of these clearly visible elements that you consider speculative when all of it is taken directly from the VMs and grounded in historical fact? But as you say, time’s a wasting, and no one should bother trying to understand the author’s intentional constructions, when they don’t make sense to you.

  42. OOTB: I’d consider the blue in the blue stripes to be *contingent* (because we don’t as yet know for sure whether or not this entire colour layer is original or added on 100+ years later), the presence of “papelonny” to be a visual motif running through the Voynich that is as yet only speculatively connected with heraldry, and the “proper hierarchical placement [etc]” to be speculative.

    And your interpretation of these is, again, speculative and grounded in speculative observation rather than in historical fact: for the other raft of historical facts that you *then* bring in to support your claims merely tells its own story, not any story about the Voynich.

  43. Out*of*the*Blue on September 20, 2015 at 8:27 pm said:

    Nick: So who slopped the paint, eh? And when? Sorry, I can’t answer that. Contingent only means the chronological situation is unclear, not that your scenario is correct. I can only point out that the specified patterns are striped and their alternating stripes are painted blue. Really and truly – without a hint of speculation. There are other issues, orientation, obfuscation, but they are resolvable without contradicting the Fieschi historical identification. Or maybe the paint was there first, and the drawing was added later. It’s no more provable one way or the other.

    Papelonny is a heraldic fur. Furs are part of heraldic tinctures; metals, colours and furs. Check out the Fox-Davies book on line. Also note the number of deficient references where papelonny and plumetty have been omitted. It is absolutely not a ‘visual motif’ scattered throughout the ms. Fish scales and roof tiles may appear to be visually similar, but there is NO intrinsic relationship. Claiming that there is one would be fundamentally flawed speculation.

    The heraldic insignia of Chotard de Chateaubriand was a papelonny pattern when he saved the life of French King Louis IX at the battle of Mansura in 1250 CE. And Louis IX was king when Pope Innocent IV fled to France. So there is a valid historical connection that would easily be known to any interested and educated chronicler of those events. But I realize you haven’t had the time and probably not sufficient interest to adequately do your homework on such an obscure topic. No worries, it’s the pun that’s important.

    So if the paired armorial insignia, bendy argent et azur, and the red galero are used to make the unique, provisional identification of the two historical figures in the rings of White Aries, then it is an absolute, clear and objective fact, that the papal figure is in a higher celestial sphere and the figure of the cardinal is in a lower celestial sphere. And this is the proper hierarchical placement for those figures and there are no two ways about it. And yet you see fit to dismiss this example of proper hierarchical placement as pure speculation. Where is the speculation? Visual analogies COULD be speculative. Positional statements are purely objective. I can see it, you can see it (hopefully), but most importantly the author could see it and that is the reason for the repeated use of positional confirmations. For example, both the blue-striped patterns are in the same quarter of the White Aries illustration. That is an objective statement of position. What is the significance of that? Well, in heraldry the upper right quadrant of the armorial insignia is considered a favored location. And if we are seeking to further test or verify the Fieschi identification, as we should be, then it is apparent that the papal and cardinal figures are both to be considered worthy of such favored placement. Otherwise, it wouldn’t matter where the were. So there is no speculation in the placement and there is no speculation in the application of heraldic standards. The only place for speculation is confined to the question of how, with all the multiple placement opportunities, … how does the illustration repeatedly contain only that singular, positional option that makes sense with the Fieschi identification? Hierarchical positioning, favored heraldic placement, the papelonny placement, the fact that the Fieschi identification is on the White Aries page and not elsewhere? And that is because only a white animal is suitable for celestial sacrifice (with reference to Homer) and a cardinal and pope in their proper celestial spheres would be presumed to have celestial connections as well, especially posthumously. And in addition the whiteness of this Aries has been subtly emphasized by the painting of the other animals and by the painting of almost everything else on the White Aries page, the most painted page in the VMs Zodiac. The manuscript is not an act of nature. Somebody did it intentionally. What comes out of the White Aries illustration, if you look at it, is what the author put into it. And then covered over with a few dry leaves.

    The interaction of the parts is too complex and focused, the number of options that don’t work is too great for this particular construction to be accidental. There is no alternative other than the intentional creation and placement of the elements I have cited within the illustration. And it is not a concocted mythology or fantasy, or something with only a single point of connection to the VMs and the rest from external sources. There are multiple elements in the VMs and each is validated by standard references, if, as in the case of papelonny or the paired Fieschi insignia, the reader happens to be cognizant of those references or is willing to learn. Just because you or I don’t happen to see it at the time does not mean that the author shares our lack of awareness. Positional placement is objective and the placement of papelonny is the author’s signature demonstration of ability and verification of identity. But first the investigator must find it and be able to interpret it as it was intended, because the specific elements are actually there in the illustrations in their necessary and proper places time and time again. Without doubt.

    So what this does, indeed, is to tell its own story. And it is a historically grounded story. So if you were investigating illustrations in the VMs, would you rather follow a line of investigation that had historical grounding or one that did not? The story is not about the VMs. It is a repeatedly confirmed, historical identification within the VMs. And what it provides is an intentional connection grounded in historical fact and the clear beginnings of a path that might help to discover a solution to the VMs. Now there is speculation for you, a speculation shared by all VMs research. But that speculation is not sufficient to invalidate the illustrated elements or their cultural and historical connections.

  44. OOTB: “contingent” actually meant “your scenario is built upon a codicological presumption which I strongly doubt, only I’m (normally) far too polite to say”.

    As for the rest: meh.

    Please stop spamming the site’s comment box.

  45. Out*of*the*Blue on September 20, 2015 at 11:36 pm said:

    A thousand pardons, Nick, I mistook you for an open-minded investigator willing to examine the evidence presented in the VMs/

    My mistake!!!!!!

  46. OOTB: I’ll be delighted to look at evidence rather than listen to speculation. Doesn’t happen much these days.

  47. SirHubert on September 21, 2015 at 8:53 am said:


    The bit I evidently misunderstood was:

    “So as to your speculating about the text, how that is done will certainly be important, but as I see it, what is speculated about, certain, specific, text segments identified by the author’s visual narrative within the VMs itself, may also be critically significant.”

    It wasn’t clear to me from this that you were referring solely to the zodiac section.

    Forgive me if I seem to be labouring the point, but when you also say:

    “The manuscript is not an act of nature. Somebody did it intentionally.”

    then that implies that you think that all the writing, drawing and colouring was performed by one individual. In fact, there are at least two scribal hands to be found, and I think I’m right in saying at least two phases during which colour was added. All that quite apart from the question of whether the text represents the original work of one individual, as opposed to something more like a compilation or epitome.

    May I also give an example where I think one of the illuminators might have made a slip? On f25v there is what looks to be a dragon (?) to the left of the plant’s roots. The illuminator has the dragon eating what appears to be a low-hanging leaf from the plant. Now, while dragons are not normally reckoned to be vegetarian, this could be some kind of visual mnemonic (it surely can’t be a dracaena plant, can it?). But when I looked at the scan on a decent screen, from what I could see of the underlying penstrokes it looked more to me as if the dragon was breathing fire, and that whoever added the colour had mistaken the flames from the dragon’s mouth for another leaf and slapped some more green on them.

    I’m sure this has been noticed before so do shoot me down in flames if I’m barking up the wrong tree, so to speak. But I do think there are good reasons for thinking at that least some of the colour was applied by a person/people who didn’t fully understand what they were looking at. For which they have my sympathy.

  48. D.N.O'Donovan on September 21, 2015 at 11:19 am said:

    I do think the plant on f.27v is meant to represent Dracaena, but with D.cinnibaris and not what we now call D.draco as the basic type.
    I do agree with you that whoever directed the ‘heavy’ painter/s was poorly informed, and less perceptive than the copyists/draughtsmen whose work is wrongly ‘corrected’ by the paint in a number of cases. I should think the work’s director, in those cases, was a Latin-trained person with more authority than breadth of learning, for he has no evident understanding of the imagery’s content – which makes me doubt that he understood the written part of the text, either.
    His aim appears to have been to make the imagery conform to the norms of Latin convention, since we find alterations to the “two fishes” which makes the end-result much closer to the usual descriptions of Pisces in standard texts such as Aratus or Manilius; again, the cruciform object on folio 79v becomes more like a Latin cross only because of the way the heavier blue paint has been applied.. etc. etc.

  49. D.N.O'Donovan on September 21, 2015 at 7:02 pm said:

    Using the Beinecke foliation, the Dracaena-and-dragon folio is 25v.
    (new scan #1006123).

  50. D.N.O'Donovan on September 21, 2015 at 7:13 pm said:

    Perhaps the dragon is engaging in a random act of mean-ness.

  51. Out*of*the*Blue on September 21, 2015 at 8:35 pm said:

    So, Nick, – I’m game. But first tell me what threshold you require for evidence.

    Is information from the historical record evidence?

    Does the fact that the VMs Zodiac has Pisces first and the fact that Pisces is a pair of fish constitute evidence?

    Can the existence of two blue-striped patterns in the VMs White Aries illustration be considered as evidence?

    If so, what do you make of that evidence? Does this actually constitute evidence in your estimation? If facts don’t suffice, what will? I’d like to get a better understanding of your perspective.

  52. OOTB: I could write an entire book on what the threshold for evidence is – the one-line version is that it’s something that can be used as a necessary part of a causative or constructional proof. But in this instance, it’s far easier to point out what isn’t evidence:-

    (1) Information from the historical record isn’t evidence. Oh, and story-telling isn’t evidence either.

    (2) No, the fact that Pisces is used as the first sign in the Voynich zodiac pages is an observation. There are a thousand parallel reasons that could all explain why this is so, and none of them is evidence either.

    (3) The patternedness of the blue-striped patterns is an observation, and a very weak and contingent one too.

    All the while you see these as ‘evidence’ rather than just observations, you will continue to significantly overestimate the strength of your assertions and claims.

  53. Out*of*the*Blue on September 21, 2015 at 9:49 pm said:

    SirHubert: If you’ve been keeping up on your reading here, you already know much of what I have to say. Sorry for the confusion of location.

    As to the identity of the author: “Person or persons unknown” will suffice for me. And as for the two ‘hands’ in the text, I have nothing to add to that discussion. Perhaps there were multiple scribes, or perhaps the construction of the manuscript continued for more than a decade and the author got arthritis, was bitten on the hand by a dog, or cut off his thumb, maybe his language became more salty, more intellectual or simply dealt with a different set of subjects. This is all pure speculation. I have no horse in that race.

    As for the dragon, do you know how many different types of dragons are flying about in fantasy worlds? They are not all fire-breathing dragons. Hard to say how far back you can push that idea. I would say that the VMs illustrations already show that the author is tricky and intentionally deceptive. If s/he thought that a fire-breathing dragon, actually breathing red fire, was too obvious an image, then intentional alterations could have been made to create ambiguity. Green fire will confuse the heck out them – speculatively speaking.

  54. SirHubert on September 21, 2015 at 10:04 pm said:

    Diane: you know what? I pretty much agree with all of that 🙂 I’m not sure to what extent the painting ‘corrections’ are deliberate as opposed to subconscious or just misunderstandings, especially if this colouring was executed considerably later than the outline. But I think that you are gloriously and wonderfully right about the link between the ignorant application of colour and ignorance of the meaning of the text. Which, at the risk of anticipating what I’ve been indolently researching for the past few months, is why I think that the letters in plants, if they are in fact colour indicators, *might* be really, really important.

  55. Out*of*the*Blue on September 21, 2015 at 11:41 pm said:

    Nick: Well, that will make short work of this conversation, because I disagree. Facts do function as evidence. For instance, the historical fact is that Sinibaldo Fieschi was Pope Innocent IV in the mid 13th century. Now whether one uses the word ‘fact’ or ‘observation’ to describe this state of affairs merely means to me that the words are being used synonymously. No need for the splitting of existential hairs. A validated observation used to validate another observation would seem to fulfill the definition.

    So all I have to do is switch my factual evidence to factual observations and I’m home free?

    I observe two blue-striped patterns in the illustration of VMs (f71r) White Aries. This is just a simple statement of existence. Why is this a weak observation? It seems to me to be as obvious as can be expected. Observation may be speculative, but if investigation produces a singular outcome, then what is that? Isn’t that a demonstration that the text contains sufficient information to match a unique historical situation?

    Suppose the VMs author really did want to take the elements I have specified in the Zodiac illustrations and create the sort of historical reference I have suggested, what more do you see that needs to be included for this construction to be considered potentially valid and intentionally created?

    Perhaps you can provide an example from the VMs that you feel best demonstrates the definition of evidence as you see it.

  56. OOTB: facts, observations and evidence are not the same thing.

    A fact is something that is, noticing it makes it an observation… but unless you can use it to causally prove something, it’s just not evidence.

    For example:
    * the little dragon has green-painted breath – this is a fact.
    * SirHubert noticed it, looked closer, and saw a mismatch between the intention of the drawing (fire) and the execution of the colour (green) – this is an observation.
    * SirHubert realised that this matched Jorge Stolfi’s pre-existing suggestion of an (original) light painter and a (later) heavy painter very strongly, and so put it forward as being part of a larger causal argument – this is evidence.

    This is not untestable speculation about authorial intent, this is an hypothesis about how the manuscript was constructed (i.e. what happened), when and how, and it has evidence that is used to support it (Stolfi had his own evidence, I put forward other evidence in Curse, but I quite like SirHubert’s green dragon breath as well).

  57. D.N.O'Donovan on September 22, 2015 at 9:44 am said:

    Sorry, can’t respond. Quite overwhelmed by the positive tone.

  58. Out*of*the*Blue on September 22, 2015 at 10:08 pm said:

    Nick: So this is really a lot of fun, but again I disagree. Here is my analogy. In a murder investigation, a gun is found, a ballistics test is conducted and a comparison is made. That comparison, either positive or negative, is an observation. That observation is a fact and that fact is considered to be valid evidence. So, in this scenario, observation becomes (is) fact and fact becomes (is) evidence. In other words, a validated observation functions as evidence. So while there may well be certain technical differences in the definitions, they are functionally equivalent, in my view, in the example.

    As to the green fire discussion, SirHubert has made an observation that the ‘fire’ is green. Acid spewed by dragons is generally depicted as green, though probably anachronistic. The number of painters involved and the sequence of paint application is conjecture. An alternative explanation is that the author simply does not want the nature of the illustration to be glaringly obvious, so he does something unexpected, green paint instead of red paint. This would be intentional deception. You may not like the idea, but I don’t think it can be eliminated as a possibility. The heraldic examples demonstrate the author’s sophistication is far greater than widely perceived.

    Intentional deception is difficult to substantiate as we certainly have many problems in identifying intentional depiction. Nevertheless, specifically in relation to White Aries, I believe it is a valid area of investigation to ask if there are potential points of comparison of the ‘quasi’ heraldic elements in this illustration with known, non-anachronistic, armorial examples in the real world. A sort of ballistics test.

    Now there are two elements in this comparison: the VMs illustration which is an image, description, observation of what is seen on the page and there is the historical example which is a fact. The comparison between the two begins as a matter of pure speculation, but, as in the example of ballistics, the similarities or differences are noted. Initial speculation leads to observation, and, given reasonable and necessary evaluation, validated observations are facts and facts clearly serve as evidence.

    So in the matter of the blue-striped patterns let us make some observations. The first matter of observation is the orientation of the stripes, and most observers seem to be naturally drawn to view the stripes within the overall radial perspective of the illustration. There is, however, a less obvious perspective to be found intrinsic in the illustration if the two striped patterns are viewed in isolation. A second perspective exists and that’s a fact.

    So, in the matter of speculative historical correspondence, you can have all the fun there that you want. I liked Belgium. What I found using the radial orientation is that a vertical pattern (heraldic paly) had a weak, probably anachronistic connection and the other pattern (heraldic bend sinister) in blue and white, showed no connections. However, using the secondary perspective of orientation (heraldic bendy) there are multiple, speculative historical connections to be made. Nevertheless, in combination with the pairing of the two armorial insignia in the illustration and the ecclesiastical red galero, the identification process yields a two factor, Venn diagram with a single, unique historical connection and the illustration provides plenty of additional confirmation. Quite a slick trick to slip the all the complex elements necessary for that extensive, complicated line of unbridled, historical speculation to exist into one little illustration, I would say. Even harder to create if you don’t know what you’re doing. Perhaps the demonstration of intent would also stand as evidence.

    What is clear is that the extensive line of historical speculation is the less obvious, secondary perspective of orientation. That observation is a fact and that is an example of deception. And not the only one. The author then has the task of clarifying the difference between deception and reality, That is done in three ways, through the objective placement of these relevant elements in the illustration to accord with standard expectations, through the infallible guidance of the pairing paradigm and through significant, valid historical connection. Assuming these factors are recognized by the investigator, that is. Otherwise someone might just breeze right over the White Aries illustration, without a single speculation, thought or observation related to any heraldic identifications and clearly many have done just that. The failure to recognize this connection of elements in the illustration is not just a missed chance at historical speculation. It is a failure to recognize a functional device implanted in the VMs Zodiac illustrations.

    In order to break a block of granite, one looks for a tiny flaw, a hidden crack. The disguised secondary orientation is at the very least a subtle pathway of speculative investigation, one that is dependent on the investigator’s knowledge of specific historical and heraldic facts. Knowledge and speculation correspond inversely. The more relevant facts you find, the more cohesive the various lines of speculation become. Focused complexity implies validity. Like a whole scourge of dragons.That seems significant to me.

  59. OOTB: never mind cracking granite, I think you’re close to breaking everyone’s will to live with these overlong comments. Good luck with your speculations, but please find another forum for them.

  60. D.N.O'Donovan on September 23, 2015 at 9:33 am said:

    I hesitate to make this comment, but before the flaming dragon item enters the list of things “which are true because my friend once said so” – Voynich land is largely made of such items – I think I should register my vote as ‘case unproven’.

    I’m still looking, but so far no comparable type of dragon pictured breathing fire has turned up. Most (but not all) the manuscript figures (so far) have either very different ears and tail, or show their teeth, or open their mouth very wide when exhaling the flames. A close example may yet turn up – till then I’m inclined to think that the dragon is just nibbling, and that the leaf is all leaf.

  61. D.N.O'Donovan on September 23, 2015 at 10:07 am said:

    update: round-eared dragons are less common, but a nice example – even with a leafy sort of foot and a leafy sort of flame(?) – in the Hunterian Psalter. 12thC England.
    MS Hunter 229.
    Part of a decorated initial.
    Can be seen at the University of Glasgow site:

  62. Is the Dragons “breath” actually breath/flame or is it simply part of the plant? It looks to me as though the dragon is eating the plant.

  63. SirHubert on September 23, 2015 at 1:35 pm said:

    Diane: don’t hesitate at all. You may very well be right – I put it forward as a possibility and no more than that. I do think that the leaf/flame is drawn a little differently underneath all the green paint than the other leaves – the lines on others seem to run parallel just about the whole way along their length, while these curve in at the tip towards the dragon’s mouth. But then again, I have no idea what the other clear pen-line running from the left-hand edge of that leaf towards the stem of the plant, starting at ten o’clock from the dragon’s nose, is meant to be.

    And non-Western dragons can breathe all sorts of things, of course. Or, of course, he might not be a dragon at all but a giraffe cunningly disguised to confuse us…

    Do we have nibbling dragons for comparison?

  64. Out*of*the*Blue on September 23, 2015 at 4:41 pm said:

    Nick: As you wish. My efforts, since I know generalities won’t be sufficiently convincing, has been to provide as much detailed, relevant evidence, oops validated observation and explanation, as will support this complex investigation. A lot of historical and heraldic information must be brought together if there is to be any chance for those unfamiliar with those topics to understand what the elements in the illustration represent. No one is required to read it – long or short. But as they say. “Whatever!” I’m sure that the whole issue can be safely ignored and VMs research can continue to spin on pointlessly for years to come.

  65. Out*of*the*Blue on September 23, 2015 at 6:09 pm said:

    So Nick, it occurs to me after my last posting that other contributors here also sometimes put out a fair amount of commentary on a daily basis, but they do it with multiple postings, with PS’s and PPS’s, and not in a single unified statement. We are all observing, speculating and commenting. I could just divide my overlong comments into little parts and not get my knuckles rapped??? Simple or complex, it matters not. Just keep it short, eh?

    PS: No current intentions to do such – just curious.

    PPS: It’s not like pretty much every simple attack on the manuscript hasn’t been tried. Perhaps something more complex, unexpected, yet innate in the illustrations presents an opportunity for further investigation of the VMs content. But if such complex matters must be summed up briefly, then it’s back to generalities, unconvincing generalities, without the relevant details.

  66. OOTB: try making one well-thought-out point at a time in a short comment.

  67. Nick: Good luck with your continuing exploration of B-408. Apparently no one has been following my translation of Fray Sahagun’s travels from Leon Province Spain to New Spain.

    An excellent source of understanding the contents of B-408 is a soft-bound book written by Miguel Leon-Portilla (and translated into English by Mauricio J. Mixco.

    You might want to add to your research of B-408 by comparing its contents with the full volume of Fray Sahagun’s “Florentine Codex/Manuscript.” written and illustrated by his two Native South American student clerks. Then you can even translate the Nahuatl which appears in B-408 — into Spanish/Latin. Also, ignore Anderson and Dibble who translated everything into English rather than Nahuatl to Spanish .
    You are only going to be able to translate the full contents of B-408 by visiting Fray Sahagun’s Florentine Manuscript.

    The very best references to Fray Sahagun’s works can be found in a book written by Miguel Leon-Portilla and translated by Mauricio J. Mixco.

    Another very interesting book I will be ordering is written by Eloise Quinones Keber, Ed. Boulder: University of Colorado Press — writes extensively about Miguel Leon-Portilla and his compilation of Sahagun’s works and publications.

    So, I am going to continue to follow their findings and discoveries — and compare the individual discussions with the various illustrations and discussions of Fray Sahagun’s magnificent ‘rough-draft’, Nahuatl-Spanish speaking Boenicke manuscript 408 (which apparently has yet to come to the interest of you, Nick, or Diane O’Donovan, or Rene Zandbergen, or Philip Neal, or Sir Hubert…)

    Mostly, I post my observations and findings on your pages here because I am doing my best to consolidate Diane’s, Rene’s, and Philip Neal’s efforts into a coherent validation that the so-called “Voynich” identification/translations were futile from the point of being a code.
    Try to start over by reading Leon-Portilla’s book.

    Although there is another excellent book by Eloise Quinones Keber:
    “Representing Aztec Ritual: Performance, Text, and Image in the Work of Sahagun” (Univ. of Colorado, Boulder). One tantalizing item (excerpt) Also, there is mention of David Carrasco’s study of female sacrifice and cosmo-magical circles.

    So, Nick, Diane, Sir Hubert, Rene, Phillip……I’ll now try to submit one more URL:


    hope it works!

    I can’t afford the price tag of $45 cloth-bound

  68. Dear Mr, Pelling,

    I was looking at some of GC’s Currier a/Currier b transcriptions of the herbal section.

    Those, and the attendant tables, seemed to show the following:

    There are approximately 9,748 words in the herbal section up to folio 57r. Most of these are on Currier a pages.

    There are about 393 words repeated two or more times on the Currier a pages, that are not on the Currier b pages. These 393 words are repeated a total of about 1685 times on the Currier a pages.

    Of these 393 Currier a words, approximately 28 end in EVA = dy, for a total of about 75 repetitions.

    Words ending in other glyphs provide the other approximately 1,610 words.

    There are about 113 words repeated two or more times on the Currier b pages, that are not present on the Currier a pages. These 113 words are repeated a total of about 505 times on the Currier b pages.

    Of these 113 Currier b words, approximately 58 end in EVA = dy, for a total of about 393 repetitions.

    The other 55 Currier b words (ending in other glyphs) found twice or more solely on Currier b pages account for only about 112 of the 505 total.

    Isn’t this large difference between the percentage amounts of repeated, exclusively Currier a and Currier b words ending in EVA = dy on the different page types the main difference between Currier a and b pages?

    Thank you.

    Don of Tallahassee

  69. Looking at the VM’s “Sunflower” I wonder if it could actually be a Carline Thistle. These are still used in traditional/herbal remedies today. They are also native to the alpine regions of central and southern Europe.

  70. D.N.O'Donovan on September 25, 2015 at 10:41 am said:

    I hesitated until I found another round-eared dragon with leafy-looking feet. Now that I’ve found one, even just one, I agree it might be possible that the Voynich dragon once breathed flame. I’ve seen other round-eared dragons (as distinct from ones with Yoda-like ears), but most were Celtic; a majority not in manuscripts, and they were even older than the Hunterian one – which I hope you liked. The Voynich dragon is by far the sweetest. Still, I tend to think it is meant to be nibbling – or even suckling since it was believed that the Dracaena fruit (on the Mediterranean D.draco) gave birth to dragons. In one of my posts about the folio, I showed an illustration from a western botanical text in which the fruits were shown with the embryonic dragons in them. 🙂

  71. D.N.O'Donovan on September 25, 2015 at 10:54 am said:

    Don of T.,
    I’ve always been impressed by the way your paradigm conforms to the patterns of the text. I’ve tried the same idea out using several other systems of abbreviated technical instructions and in my totally non-cipher qualified opinion, it does seem to be just the model we are looking for. In a sense, an artificial language, but not having any simple plain-text prose or poetry as its basis. Just thought I’d say.

  72. The South American ‘dragon’ was a large spiky lizard called a Gila Monster. I had one as a pet. If one were to lay the lizard on its back, it would think it was dead. The myth of its bite being poisonous is false (but it can hurt like a pinch).
    “Somewhere else’ in the VMS there may be mention of the Monitor lizard also. I can just see Sahagun’s clerks trying to distinguish beteen the two lizards: Gila Monster or Monitor?

  73. Nick, what criteria do you use to break down the text into the smallest section that can be studied on its own for these purposes? Is it even possible to take the botany pages and study them without constantly having the urge to try and relate to the rest of the work?

  74. boyfriend on September 27, 2015 at 7:44 pm said:

    Who will be the Champollion of Voynichse ?

    Of course I know the winner. As it is, it´ll be me.
    I know the key.
    It is written on the side of 116. It is also, of course, written on the first page of the manuscript. And what is very interesting, as is written on the cover letter, which is in the Beinecke.
    On the cover letter is written guidance on the translation.
    Exactly the same instructions as on the side of 116. Look for the logo Voynich. Cat playing with a mouse.

    The first line on the cover letter :
    Czech book, substitutions, 1,2,3. ( Numbers 1,2,3, refer to the gematria ).

    Michael Hadbank Voynich, and his wife, knew the instructions ( algoritmus ).

    The entire manuscript is written in the style : The author plays with you like a cat with a mouse. ( logo Voynich = Beinecke, cover letter ).

    Your fan and friend.

  75. D.N.O'Donovan on September 28, 2015 at 12:42 am said:

    “the cat who brings the mouse” appears to be a metaphor for the merchant, or perhaps more generally the low-status agent who nonetheless gets things done. It appears as a motif in the Palazzo Datini, in the works of the Venetian Michael of Rhodes, and in other similar contexts. I expect its key is a proverb – though in Italian medieval proverbs the cat is a female.

    One possibility is the more sober section from Bartholomaeus Anglicus, Bk 18 of his De proprietatibus rerum; [The cat] “He is a full lecherous beast in youth, swift, pliant, and merry, and leapeth and reseth on everything that is in front of him: and is led by a straw, and playeth with it; but in age he is a right heavy beast and sleepy, lying slyly in wait for mice: and he is aware where they are more by smell than by sight, and he hunteth and reseth on them in privy places: and now when he takes a mouse, he plays with that, and eats it after the play.”

  76. boyfriend on September 28, 2015 at 1:52 pm said:

    Dear Diana. And friends.

    As it looks, I´m my comment misunderstood. Commenting writing that Michael Habdank Voynich and his wife knew the code to the translations of the manuscript. Code and instructions on translations of the manuscript is written on the cover letter. Who is at Yale. ( Beinecke library). There it is written : Czech book . This means that the Voynich and knew the instructions for translation.
    On the cover letter is written instruction to decrypt.
    Instructions for the translation it self is exactly the same as it is written on the page 116 of the manuscript. And it is the same, so it proves that the Voynich known algorithm. She knew him and his wife.

    Quite clearly is on the cover of the letter reads : Czech book…….and instructions. The instructions are written in the Czech language. When writting a letter is used for substitution.

    A friend of all people.

  77. Tyler: I’d recommend Herbal-A pages, Herbal-B pages, Quire 13, and Quire 20 as the four basic sections for Voynichese analysis. Each has its own idiosyncracies and patterns: and much as many people are tempted to think that using a larger corpus will get you to the finishing line quicker, I strongly suspect that smaller will prove to be significantly better in this instance. 🙂

  78. To add to what Nick has said, Quire 20 is excellent for those wishing to study the text alone. As it has no diagrams or drawings it is impossible to infer incorrect meanings from them. In other parts of the manuscript it is easy to be swayed by your thoughts on the pictures.

    Also, Quire 20 is fairly uniform in terms of text statistics, and shows strong paragraph and line based patterns.

  79. Nick & Tyler: A good source for understanding the language being used in the “Voynich” is an on-line edition of the entire contents of the so-called “Florentine Manuscript”.
    If your computer has a ‘read-only’ feature, you will immediately be able to compare the illustrations AND side-by-side Latin and Nahuatl columns of text which identify every illustration of animals, herbs, sun and moon legend,
    goddess of the lake and many other gods and goddesses, mushrooms and toadstools, snakes — poisonous or not, festivals, birds of all kinds including water birds for their plumage/feathers — and birds which were used for fishing in the lake. Someday, I may post my ID’s of the silkworm and other butterflies AND moths which Sahagun discusses.
    The very last pages of Fray Sahagun’s enormous book
    illustrates (with little dialogue) his scribes making paper (under a mulberry tree which is being strangled by a wild fig tree/vine). The manuscript shows two scribes writing and illustrating on the paper (supposedly made from strangler-fig tree-vine). But elsewhere in the manuscript Sahagun discusses making paper from mulberry-tree bark.
    Concerning all of the floral and bush illustrations and discussions, one can compare Yerbas buenas with their various uses as medicines and/or beneficial health maintenance herbs.
    There are also many pages of illustrations and accompanying dialogues for the many types of, and uses for, the corn plant , its accompanying squash vines, and bean plants — all growing together.

    Another interesting aid to translating the dialogues throughout the so-called VMS (B-408) Was Fray Sahagun’s Psalmodia (Latin and Nahuatl ) ‘songs/prayers/saint’s days.

    So, I no longer ponder. I have translated some 40 folios. The first two I identified and discussed were the yucca/agave, and the tomatillo. Two other plants I’ve pointed out as having their colors and discussions exchanged appear in the same folio. I’m pretty certain several particularly eagle-eyed botanists will be able to relate to the radicchio and the cilantro/coriander.
    I have neither the strength or ambition to be the “Champollion of Voynichese”.
    Good luck, y’all!

  80. Diane – Thanks for noticing I am trying to let the glyphs tell me what they mean, rather than requiring any proposed solution to obey preconceived ideas, answer preformed questions about perceived problems, use officially sanctioned methods or be found only after computing some algorithm on a computer using some unheard of language (all for a system invented in 1421, and one that has been indecipherable using those same methods for a hundred years). What is it they call it when someone keeps doing the same thing, expecting different results? -Don of Tallahassee

    “Never worry about theory as long as the machinery does what it’s supposed to do.” -Robert A. Heinlein

  81. SirHubert on September 30, 2015 at 8:10 am said:


    “What is it they call it when someone keeps doing the same thing, expecting different results?”

    I’ve never quite understood why the conventional answer to this is “madness.” It seems to me that “practice” is an equally good answer.

  82. SirHubert: I thought the answer was “dating”. 😉

  83. Interesting paper – thanks Rene.

    I think it’s a bit of a stretch to change a Siberian tiger with a sable into a domestic cat with a mouse, but connection to the earlier Venetian printers is certainly reasonable. Doubtless the name struck a chord with Wilfrid.

    The earlier period also seems appropriate; most versions that I’ve come across (as symbol of the canny and successful merchant) belong to the fourteenth-to-sixteenth centuries. It must have been quite an established trope by the time the Sessa printers made it their mark.

    “The Sessa used a charming printer’s mark, variously elaborated, depicting a cat with a mouse in its mouth.”

  84. boyfriend on September 30, 2015 at 2:59 pm said:

    Friends and R. Zandbergen.

    As seen so still you do not understand what you write. I will show you graphially.
    The first line on the cover letter.

    Concernine ( q) the Cipher MS.

    Voynich manuscript substitution . ( Gematria ).

    Swap characters :
    Czech knihe ( a) the cipher 1.2.3. ( 123 = MS).

    Sing the letter ,,M,, is composed of two numbers.
    ( numbers mean = gematria ). ( Character letter ,,S,, has a volue numbers = 3 ).

    Concerninq = Conce ,, rninq ,, = Czech ,, kniha ,,).
    ( kniha – czech language = book – English ).

    Ethel and Michael are playing with you. ( The cat and mouse). Ethel studied Slavic languages. ( czech = Slavic ).

    I will write you again gematria.
    1. A.I.J.Q.Y.
    2. B.R.K.
    3. C.G.S.L.
    4, D.M.T.
    5. E.H.N.
    6, U.V.W.X.
    7. O.Z.
    8. F.P.

    In the second line is written guidance on the translation. In the Czech language.

    And because I know key ( manual ) so I Champalione.
    Instructions ( manual ) are the same as on page 116 MS.

    Meaning the first line :

    Czech book. Gematria encrypted.

    Your boyfriend.

  85. Nick & all (u too Fumblydiddles): Over the past two years I’ve given you the Latin-American ‘alphabet’ and syllables as used by Fray Sahagun’s clerks and artists. The “Voynich” began as Sahagun’s brief diary, and his subsequent collaboration with students at the Colegio in Santiago, Mexico.
    All of Sahagun’s works (including his diary and notes) were confiscated by the Spanish Inquisition — and never returned to him before his death.
    So, several centuries later, Voynich bought the rough draft manuscript from a derelict Papal library/store-room. We’ve been puzzling over it now for almost a century .

    No code. Spanish/Latin/Nahuatl would spell the word Voynich as: ‘ eu-oi-n-ch ‘


  86. PS: If any of you have the credentials or education to approach Professor Miguel Leon-Portilla, you may be quite gratified in your quest for a “Champollion of Voynichese”.
    Senor Portilla’s, and translator Mauricio J. Mixco’s, biography of Fray Sahagun is a fascinating look at earliest Latin American life consequential to the European invasions.

  87. boyfriend on September 30, 2015 at 7:56 pm said:

    For pale blue eyes. I will write it in numbers.

    Not difficult to understand. 🙂



    C..Z..E..C..H….K..N..I..H..A.. 🙂 ( czech language )


    Czech book . ( english language )


  88. Dear Beady,

    The Nahuatl words look too long, on average, to match up with the VMS words. Or are you envisioning a one VMS glyph/one Nahuatl syllable congruency for some/all VMS glyphs.?

    Do you have a transcription alphabet and any translated/transcribed Voynichese into Nahuatl pages (similar to my attempted decoding pages)?

    That way I can check the words against any dictionaries the Nahuatl may have left behind.

    One other question – did the Nahuatl have marijuana plants?

    Thank you.

    Don of Tallahassee

    “I’ll try anything once, twice if I like it, three times to make sure.” ― Mae West

  89. I have just about done a full circle on the VM. I am now back at Basque/Iberian. The ‘Apothecary jars’ look like ‘Albarellos Farmacia’ to me (Google image search).

    (I am not sure if it’s entirely relevant to the VM but Ignatius of Loyola the founder of the Jesuits was born in the Basque country.)

  90. Dear Beady,

    I’m obviously missing something important.

    Why would the Nahuatl use the name Voynich (eu-oi-n-ch)? Or spell it?

    Can you show or tell me where in the VMS the name appears?

    Thank you.

    Don of Tallahassee

    “There are no insoluble problems. Only time-consuming ones.” -James A. Michener

  91. bdid1dr on October 1, 2015 at 3:31 pm said:

    If instead of repetitions of a certain phrase or word, you can find a number for certain measures/amounts of any particular ‘pharmaceutical or recipe, please let us all know.
    aqa aqa aqa + ixtomatillo = ixtomatl ‘juice’ ‘drink’ ‘pulp
    uno tomatillo + tres aquas + tomatillo juice

    Yucca/agave leaves and roots, mashed into pulp, and juice, make soap and shampoo.

    Green cilantro leaves and chopped tomatillo make salsa.
    Cilantro plant produces seed for spice coriander.
    On the same folio can be found a red and white ‘cabbage’ called radiccio or radicchio .

    The problem with recognizing and relating the dialogues to the illustrations is simply because the cilantro was painted red, and the radicchio was painted green and white.

    Yes, Dan, Iberian is close enough to the monastery in Leon Province which sent a group of young monks from Sahagun Spain to “New Spain”.

    Cactus ‘tunas’ (nopales) are fruit which can be eaten (once spines have been rubbed off. The ‘tunas’ can also be mashed and the juice fermented for pink or red dye-product..


  92. boyfriend on October 1, 2015 at 10:00 pm said:

    Blue eye.

    I will not write about the tomato. It is my reget tomatoes in the manuscript are not. But I will show you different numbers. Page 2r. Numbers 3333. ( C,G,S,L.)

    I will show you what a symbolic plants. 🙂

    Instructions for decryption is written on the first page of the manuscript.

    On the second page 2r.
    It is drawn symbolic plant.

    Important here is the root ( ROOT) !!!
    Root is meant as a base. Grows from the root = word. !!

    The root is composed of the letters . ( C.G.S.L ) !!

    This author showed you gematria !!

    Which is used for handwriting .( manuscript).
    Root = foundation. ( basic).

    From the root – the letters – there is a word.
    The word is the basic for the sentence

    The author writes that sows letters. !
    Plants were grown from seed. !
    From seed grows a root.
    The root of plants (word ).

    Therefore you author shows the second page of the manuscript.

    Root = 3333. = Gematria. ( C,G,S,L ). 🙂

  93. boyfriend on October 2, 2015 at 7:38 am said:

    Blue eye

    I use a translator, so sorry. Repair.

    The author writes in the manuscript. Not that he writes.
    But that sown ( SOWN) characters letters.
    He writes therefore means sown letters.

    ( sown = writes ).

  94. Dear Boyfriend,

    I have looked at the image of the roots on f2r.

    I don’t see the letters (C, G, S, L). I don’t see (3, 3, 3, 3).

    If I try very hard, I can see a Greek column top or a stick figure with huge boxing gloves or a two legged Mr. Krab. I can even imagine that two (angry?) eyes and the top of a nose are to be seen,

    If letters are to be found, I can possibly see (P, I, I, G) or
    (P, L, L, G) or some combination thereof.

    I do not understand how you arrived at Root = 3333. = Gematria. ( C,G,S,L).

    Just like you, I don’t see any tomatoes, either. I regret it, too.

    Would you please explain again to help me understand?

    Thank you.

    Don of Tallahassee

    “If you don’t ask the right questions, every answer seems wrong.” Ani DeFranco

  95. bdid1dr on October 2, 2015 at 3:20 pm said:

    Magic words: Tomatillo, Yucca, Mescal, Mujeres……..

  96. boyfriend on October 2, 2015 at 5:27 pm said:

    Don of Tallahassee.

    Symbolic plants ( page 2r).

    Root = is composed of the letters.
    Root = is composed of the C.G.S.L.

    Letters C.G.S.L. = gematria numbers – 3.

    Author showed you gematria. When handwriting is used substitution cipher. Based on gematria. In the middle ages, the substitution is commonly used. ( This is a very complicated encryption.)

    You got at least a word : Concerninq . ??

    The manuscript was written around here. In Bohemia. ( Czech).

    The whole is written in the Czech language.

  97. Dear Boyfriend,

    I am sorry. I do not understand gematria, how you get the root letters or how you transpose one into the other.

    I only know two words spoken in Czech, none written.

    Concerning is an English word – is concerninq a Czech word? Are you conflating the two words?

    I think I do not understand the part about the author showing me gematria.

    If you are correct, then 3333 means Czech book? Do I understand this correctly?

    I admit to being befuddled. Sorry.

    Thank you.

    Don of Tallahassee

    “The obvious is that which is never seen until someone expresses it simply.” -Kahlil Gibran

  98. boyfriend on October 3, 2015 at 2:52 pm said:

    Dear Don Tallahassee

    Do you know at all what it means substitution ???

    Thinking means shit to know. I mean a lot. Shakespeare.

  99. bdid1dr on October 3, 2015 at 4:06 pm said:

    Follow up on what Don Hoffmann recently posted herein.
    I used the syllables ‘oi’ or ‘ue’ simply to x-pl-an that there is no “V’ in the so-called “Ou-o-i-n-i-ch” manuscript.

    As far as reading the contents of B-oe-ni-ke man-u-s-cr-ipt
    408: look for and compare the combined syllabic glyphs which look like very elaborate “P” s. They most often begin a new Par-a-graph (all in one long stroke of the pen) of discussion.
    Another very interesting glyph is what Nick called “Brackets”
    The “Brackets” are actually the stretching the “l l” glyph over a group of letters and reinserting the ‘l’ farther down the sentence:. Example: l-o-ca-l-i-t : (English locality)
    PS: The tl glyph doesn’t work that way. tl is the major symbol by which the language can be identified. So, the word locality, in New Spain, would appear as ll-o-ca-tl .

  100. boyfriend on October 3, 2015 at 8:06 pm said:

    Pale blue eye.
    I must commend you. You once wrote here about the picture on which the two eyes. At the top of the page. Sentence is written in Latin alphabet. Part of the sentence is crased. ( deleted) I translated the sentence.

    It is written in czech language.

    Part of the sentence is : Slezska the beard has muńk…….Then it is deleted.

    Muňka = czech language.
    Crab louse = english language.
    ( Beard = hair). ( beeatles on pussy). ( pusy = (I).


    Image has two meanings. Eyes and pussy.

    Otherwise , a Mescal in manuscript writing.
    The tomatoes did not.

  101. Don: hopefully that answer is enough to convince you to stop feeding the bear(d)s.

  102. boyfriend on October 4, 2015 at 9:01 am said:

    Nick and Don.
    I like to help everyone. But when Don knows no substitutions. So it is wrong.
    When handwriting is used substitutions. Homophonic substitution.

    2. For more difficult to decrypt the text inserted ” Deceptive letter “.
    What it is, it is written in the manual. ( page 116).

    3. And what is important . When the decrypt text is shifted characters.

    Sample : isshi ftedc haract ers.

    4. At the same time the text. He must finish characters. For a specified character. It is also written in the manual ( page 116).

  103. bdid1dr on October 4, 2015 at 3:33 pm said:

    The item in B-408 is a to-ma-tl -o : (a ‘husk’ tomato):

    The tomatillo has a distinctive husk covering which eventually turns dry and white — and the husk eventually splits open to reveal the greenish/purple fruit which ripens into red-purple.
    There is NO mention of a similar, but much larger, fruit which Senor Zandbergen refers to as a tomato. You will not find mention of the larger fruit (which has NO husk) anywhere in B-408.
    Even today, scholars everywhere stick the tomatillo into botanical discussions of the gooseberry (a distant relative more often used by early settlers to East Coast North America.

  104. bdid1dr on October 6, 2015 at 5:01 pm said:

    I further refer you to your Cipher Foundation page. NO, I do NOT want to be the “Champollion-of-Voynichese” ! Which is why I’ve most recently been commenting on your “Foundation” pages.
    Good luck, Diane, Rene, Ootb, Sir Hubert, Rich Santa Coloma (who just never ‘gets it’), and ProfZ (who may be just a little ‘zany’, but tries hard to be coherent).

  105. bdid1dr on October 6, 2015 at 5:19 pm said:

    Furthermore: Who invented/designed the EVA (Electronic Voynich Alphabet)?

  106. boyfriend on October 7, 2015 at 7:27 am said:

    Pale Blue Eye. 🙂

    You do not want to be Champollione ?? 🙂

    Strange. Too weird.
    But that´s good. Because Champollione am I.

    ( Ty jeden snaživej blázínku ).

  107. bdid1dr on October 8, 2015 at 3:06 pm said:

    Come to think of it, who is/was Champollione? Thank goodness “Boyfriend” has come to your rescue!

  108. bdid1dr: “Champollione” started out as Jean-François Champollion but then got lost in the translation. Which is, of course, richly ironic. 😉

  109. boyfriend , Champollion,,. :-) on October 9, 2015 at 10:10 am said:

    Pale blue eye + friends

    I like to help everyone. I´ll help you to day.
    Watch how simple it is.
    The second line, the cover letter. ( letter Ethel + Voynich)

    Beginning three word .

    I Not to be = english language
    I Not tobě = czech language.

    Word , Not = Nota…in the Czech language means = instruction.

    Word : tobě ( english – you ) are demonstrative pronoun.

    Also look at how the character is written ,, N,,.
    It is written as a letter ,, C ,,.
    So when you read the letter ,, C ,,. Instead letter ,, N,,. So read word ,, Cot ,,.
    Substitution No.4 = D,M,T. ( Cot = Cod ) , = ( Code ).

    Ethel studied Slavic language .

    The beginning of the second line, that is :
    I Cod ( Code ) you. (( I Cod tobě = Czech language).

    What word ,, cot,, of course, it has another meaning. And in the Polish language. ,,Cot ,, which means in Polish language ,,Cat ,,.

    A cat has Voynich = the logo. ( Mouse + cat ).
    Michael = Michal , diminutive name is Misha.
    ( Misha = english language )
    ( Mišo = Slavic , Czech , language ).

    In the name of Mišo. Is hidden mouse. Just sing replaces the letter ,, i ,, ( i = y ) substitution.

    Mišo = Myšo = Myš . ( Myš = mouse ).

    Mišo Habdank Voynich logo, like a mouse. The cat is the code.
    Cat playing with a mouse. Means. Voynich with you playing.
    Ethel and Mišo knew the code of the manuscript.

    But Voynich wanted more money. So he tried to cram into the manuscript Bacon.

  110. bdid1dr on October 9, 2015 at 4:48 pm said:

    Of course, how could I forget the name of the ‘translator’ of the Egyptian hieroglyphs ? !
    I tell ya, Alzheimer’s is closing in on me !
    Has anyone else ever noted that the pyramids were aligned with the star formation “Orion’s belt” ?
    Cross-refer with the activities of Zahi Hawass and the robotic camera which brought beautiful “live” photographs of the great pyramid’s interior corridors and doors.
    I wonder what Champollion’s reaction would have been if he had been given access to the tomb’s interior.
    ;-^ (tongue-in-cheek, as usual)

  111. bdid1dr on October 12, 2015 at 1:32 am said:

    Prof Z (boyfriend): There is no tomato in either the “Voynich” manuscript nor Friar (Fray) Sahagun’s tremendous book, which is now called the “Florentine Codex” or the “Florentine Manuscript”.

    The Voynich manuscript (which was Fray Sahagun’s diary and prefix to Fray Sahagun’s “Florentine Manuscript”) does have pictures of both the tomatillo (physalis ixocarpa) and the Yucca / Agave . Keep on looking, and keep on translating into Latin (only if you are able to read and translate into Nahuatl and Spanish, instead of Czech, or Hungarian, or Polish, or Hebrew).
    You can also find a a horned toad, instead of a dragon, in the Voynich Garden (and in the Florentine Codex/Manuscript).
    I admit, I do have blue eyes, but they now have cataracts which cannot be surgically corrected.
    So, this is why I call myself ‘beady-eyed’ wonder-er

  112. bdid1dr on October 12, 2015 at 1:37 am said:

    PS: Anybody been able to find an armadillo anywhere in the “Voynich”. Just ‘wondering — there is one in the “Florentine Manuscript, Part 12: “Earthly Things”.

  113. boyfriend , Champollion,,. :-) on October 12, 2015 at 12:42 pm said:

    Dear Bdid 1dr.

    I´m sorry, of course. I have a problem with your eyes. If you lived closer so I could help. I am a good magician. In a few minutes I will heal.

    Otherwise, the animals in the manuscript drawn.
    In the Midle Ages in Bohemia. Much of animals. Every nobleman have brought what he needed. Zodiac also had a great nobleman Rosenberg ( Czech language – Rožmberk).
    He was also. of course, a large botanical garden. He also had greenhouses. Which grew everything that available at that time. Rosenberg was the first of the king. His daughter Eliška ( Elizabeth) is the author of the manuscript.

  114. bdid1dr on October 13, 2015 at 3:41 pm said:

    Diane & Rene, have you had the opportunity to compare the “Florentine Codex” contents and illustrations with various mysterious botanical illustrations and accompanying nomenclature? (Chapter 11 is particularly generous with discussion of the ‘yerbas buenas’).
    Here in California (San Francisco) there is a whole neighborhood called “Yerba Buena”.

  115. Gregory on October 14, 2015 at 8:09 am said:

    chłopak, Champollion ,,. 🙂 – dupová vyčeračka.
    Strč prst skrz krk

  116. bdid1dr on October 14, 2015 at 4:34 pm said:

    The Yerba Buena neighborhood, has in more recent years, become famous for its park, art galleries, and museums.
    San Francisco’s “Mission District” (at one time considered to be a slum) has ‘blossomed’ also. I hope to re-visit San Francisco once more before winter sets in.
    Actually, San Francisco is famous for its weird seasons: July, August, and September can feel like winter, while waiting for a bus or streetcar at Van Ness and Market Streets.
    While waiting for buses in other parts of ‘town’ or wandering Golden Gate Park, I would encounter Jerry Garcia and his band members passing by me on the stairs to the old Victorian house (practice sessions). And then there was the day I met the Hari Krishna free food distributors and dancers. Don’t get me started on Janis Joplin (Sheeps Meadow Stage in G.G. Park) —- o-o-o-o-o-whee, screaming meemee ! I couldn’t understand a word she was screaming. How’s that for reading lips or translating a single word?

  117. Gregory on October 21, 2015 at 5:14 pm said:

    What do you think about that: the biological part Voynich Manuscript should be understood in the same way as described by Albert Barillé in the French animated series Il Etait Une Fois … La Vie – Once Upon a Time … Life.
    In the blog http://gloriaolivae.pl/ I present my interpretation.

  118. bdid1dr on October 23, 2015 at 4:08 pm said:

    Finally — it appears that several tribally-unrelated Medicine Men are collaborating in the creation of an ‘encyclopedia” of traditional medicine/cures.
    So, is that the case? Or is it just one tribe’s efforts at getting their elders stories, legends, and cures into WRITTEN/PRINTED/ILLUSTRATED form and made available to researchers/university archives/general populations ?

  119. bdid1dr on October 24, 2015 at 4:24 pm said:

    Correction to my post in re Book 12 of Florentine Codex: It is Book Eleven which discusses AND illustrates “Earthly Things”. Everything from ants and termites to ‘yo-ll-o-to–to-tl” .

    Because Anderson & Dibble translated the Nahuatl/Spanish discussions into English — we still don’t get a clear name for the bird. But if one reads the dialogue/description, one might come to the same conclusion/translation/identification of the bird: ‘mourning dove’. Sound: ooh-hoo-hoo-hoo !

    My recent visit to our local botanical sales garden was very interesting; she had a talking parrot in a large cage; and a wild bird in a nearby tree (not a weeping willow tree). The proprietor was able to identifying the ‘weeping” bird as being the mourning dove.

  120. bdid1dr on October 24, 2015 at 4:35 pm said:

    PS: The recent 78,000-acre wildfire burned to less than a 1/2 mile (or 1/3 kilometer) of our mountainside home. We are blessed!

  121. bdid1dr on October 28, 2015 at 3:49 pm said:

    @ Philip Neal and Donald Hoffmann:
    My apologies for skimming right by your offerings to this latest discussion.

    Mr. Neal, it was material which I found in your huge archive that sent me down the paths toward an obscure Papal (Gregorian) documents/library (Fray Kircher), Suleiman, Busbecq, Rudolph II, Alban Lake and Nemi Lake, Frascati. Onward from there to catch up with Nick and his friendly, long-time correspondents.
    I hope you are following the discussions in re Fray Sahagun’s enormous archived (and published recently) “Florentine” manuscript.

  122. bdid1dr on October 28, 2015 at 11:44 pm said:

    Especially Book Eleven “Natural Things”
    The so-called “Voynich” manuscript is loaded with discussions of the writer’s homeland, it’s lakes and monasteries, the Spanish province of Leon, his education, his journey (by sail) to the “New World”. The rough drafts of his visits with the elders of the many villages surrounding Lake Titicaca.
    Much of the “Voynich” manuscript’s notes and illustrations ended up in the “Florentine Codex”. However, there was a serious interruption of Fray Sahagun’s written works; when a Dominican priest reported him to the Spanish Inquisition.
    Eventually Sahagun did get to finish his 12-volume Codex.

    Apparently Fray Sahagun never did get his diary/rough notes returned to him. He died not long after enduring the Inquisitional trials (and was buried in Santiago/Mexico City.

    Curiousity killed the cat; satisfaction brought it back. I still think the so-called “Voynich” manuscript was bought (by Mr. Voynich) from a closed-down/out of business Papal Inquisitional archive. I am so very pleased that the Boenicke Library has given the so-called Voynich manuscript a number, at least: Boenicke Manuscript 408.

  123. bdid1dr on October 28, 2015 at 11:54 pm said:

    PS: My favorite translations of B-408 folios are the “Mushroom” and “Mulberry”. Some others are “Scabiosa” and “Plantago Ovata”.
    More recently, I’ve been identifying and translating folios which discuss various birds and insects (mourning dove, ants, flies, termites, hornets, and sericine producing butterfly. to name a few (Nahuatl-Spanish-English).

  124. bdid1dr on October 29, 2015 at 12:04 am said:

    I’ll leave discussion in re valuable minerals and stones to Nick, you, and other interested bloggers.

    Nick: Have you yet found any discussion in B-408 for corn and squash? The “Voynich” does have an illustration AND discussion of the “Oshquash BLOSSOM” .

  125. bdid1dr on October 31, 2015 at 4:24 pm said:

    Halloween is impending. Pumpkins (members of the Squash family ) will probably appear ‘here and there’ regardless of the ‘homelessness’ of several hundred families who lost their homes in the most recent (78- thousand-acre) forest fire.
    beady-eyed “witch” – just for tonight .

    B-D-I-D W……. 🙂

  126. bdid1dr on November 2, 2015 at 12:49 am said:

    No ‘trick or treat’-ers visited us last night (on our perch on the mountainside at the 3,000 foot elevation). Can’t blame them for not attempting the climb up the road (with 3 and 4-year old children in tow). I did buy a couple of bags of candy, and left them on the counter of our re-opened (post-fire) grocery store and delicatesson — for the proprietors to distribute.
    Hope y’all had a ‘scary enough’ Hallowe’en !

  127. bdid1dr on November 3, 2015 at 4:45 pm said:

    How to identify the ‘silkworm’ butterfly/moth (papillon) :
    In the ‘Florentine Codex” there is an illustrated discussion of the butterfly and moths; only one stands out because of the cocoon which is pictured next to the butterfly. The artist was attempting to illustrate the importance of that winged insect: the silkworm butterfly (sericine producer for the silk industry.
    In another section (of both the “Voynich” AND “Sahagun’) one can find discussion of the leaves (mulberry) being fed to ‘blattae” (larvae) — until they spin a cocoon. Before the moth/butterfly can emerge from the cocoon, the cocoon is tossed into boiling water.
    So, if you feel like following up on this comment, first look for that very strange illustration of what ‘looks-like’ a pineapple in Boenicke 408. The illustrated item is a single MULBERRY FRUIT. So, one can then find discussion about the tree’s LEAVES becoming ‘porridge’ for ‘blattae’ (infantile silkworms).

  128. bdid1dr on November 14, 2015 at 5:09 pm said:

    Terminology correction: “pabulum” — “pablum”
    Sericine is another term or ‘nickname’ for Chinese development of the Silk industry. “Top Secret” for many centuries; until imported to Europe via hollowed out ‘canes/walking sticks.
    Nick, I’m pretty certain you’ve read my earlier donations in re silkworms and their eventual arrival in European territories.
    bd 🙂

  129. Anton Alipov on December 15, 2015 at 6:25 pm said:

    A question about the gallows. Are there any occurrences of “simple” (as opposed to benched) gallows as either:

    – the last symbol of the paragraph (or the entire folio);
    – the last symbol of a standalone label.


  130. Anton: word-final gallows are certainly rare, but I’ve never tried to quantify the rarity of paragraph-final gallows or label-final gallows. I would be surprised if either type was more frequent than word-final gallows.

  131. Anton Alipov on December 16, 2015 at 2:02 pm said:


    According to the count in the “Voynich Reader” tool, word-final gallows are rare indeed, with k having less than 90 occurrences and p, f and t having even less.

    However, I cound not find offhand an easy way to obtain stats for paragraph-ends and label-ends specifically. Suppose that there are no such occurrences at all. This given, we would have the following picture for the set of the basic gallows.

    – They have not so many variations – just four (p, f, k and t).
    – They are not rare in the corpus.
    – They occur very frequently as the starting character of the paragraph. Sometimes many paragraphs in a row begin with the gallows.
    – They never occur in the end of any distinct high-level logical entity (paragraph or label).

    What set of elements could have such properties? A set of articles!

    E.g. (modern) English has two articles – “a” and “the”, German has five (der, die, das, ein, eine) etc.

    Just an offhand idea. Was it ever considered or put under test?

  132. Dear Mr. Pelling,

    You seem to admit in the post in your blog that you cannot distinguish between the glyphs well enough to admit that my proposed Voynich Lite deconstruction scheme does anything except “works well: but – in my opinion – it doesn’t yet help us move towards what I would consider any of the basic milestones we would need to pass before we can crack the puzzle of Voynichese.”

    You then go on to give a lot of reasons that seem to make the process an almost impossible one because you cannot tell what the different glyphs really are. You say you must be able to read, parse, tokenize and model before the process can be accepted by you (or that it may be impossible to attempt even before the process can begin because one or more of the four requirements of yours cannot be met beforehand).

    In the proposed Voynich Lite (available at my Fumblydiddles.com site), all I try to do is show a systemization of the glyphs and their positions within the glyph words. No code meanings, no word meanings, no definitions, no parsing, no tokens, no reading – just a system and pattern – one that (in your own words), “works well”.

    Is it really true that you feel a possible way to show how the VMS words are constructed/deconstructed (one that seems to work well for 503 of the 505 most common words in the VMS) is unimportant and not worth further comment? To my knowledge, nobody else has done anything like this with Professor Stolfi’s ideas. Or the VMS words.

    (And all I would need to do to get the remaining two words to fit is add one one-glyph code – which I won’t do just to make it all come out perfect, because I think the two words are exceptions to the rules, as are some other less often (less than 10 times) repeated VMS words.)

    Is it too simple a system? Were you hoping for more? Does it give results you want to disagree with for some reason? Is it not consistent as all get out?

    The 505 words are each repeated ten or more times in different places in the VMS. Together they comprise about one half of all the glyph words in the VMS. I am showing how half the words in the VMS can be deconstructed with way less than two hundred code groups. And they can be constructed/reconstructed using the same sequence in reverse.

    The proposed system, tables and sequence shown at my site (of which the Voynich Lite is only a sample), when extended to all the other words in the VMS, seem to work for all but a small percentage of all the words in the VMS. And except for the first large group of codes in Group I (or Table I), I don’t need many codes to do it.

    Is there any way I can get you to give my proposed deconstruction ideas a second consideration? How about money – if I sent you some, would a trip to the pub (you reference at least one establishment as a meeting place for group sessions) be likely to put you in a more accepting mood? (Nudge, nudge, wink, wink – say no more.) : )

    Send me a postal address, Swiss or Cayman Island bank account number, or drop zone coordinates, date and instructions. I will send you all the halfpennies, fivers and farthings I can amass.

    Thank you. 🙂

    Don of Tallahassee (Voynichero)

  133. Don: what I tried to get across in this post is that I would categorise your current approach to the problem of modelling as being far closer to data collection than to actually understanding the low-level principles that drive the (large) set of Voynich words. Collecting together the Voynichese dictionary doesn’t help us explain whether “4o” is a letter, a word, a prefix, a null or what: and it doesn’t help us understand why some “word parts” (e.g. “ol-” or “l-“) occur much more in some parts of the manuscript than others, let alone grasp how Currier A maps onto Currier B etc. The point of modelling is to try to answer those kinds of questions, and no amount of sponsored beer will change that basic position. 🙂

    PS: good luck with the elections! 😉

  134. Dear Mr. Pelling,

    I am exactly trying to show the results of data collecting and integrating those data to show how the words in the Voynich Manuscript can almost all be constructed or deconstructed using one system, one set of code groups and one sequence for doing things.

    Doesn’t showing how the words have structure beyond Professor Stolfi’s prefix-midfix-suffix ideas mean anything to you?

    My proposed solution does show how almost all of the VMS words can systematically be taken apart into smaller parts using the same system, code tables and sequence – in the case of the Voynich Lite proposed deconstruction of the 505 most common VMS words, with a remarkably small number of codes in all but the first group.

    It doesn’t try to answer higher grade questions or explain your preconditions – it shows facts.

    Is this all just a happy accident? How did that happen? How can I make the glyphs obey my proposed rules in so many VMS words.

    The proof is all right there. I’m not making stuff up.

    My Voynich Lite really does do just what I say it does. You, yourself, admit it works well in the second paragraph of this thread.

    You have complained that Professor Bax has decoded too few words and now seem to reject the fact that I show how to deconstruct too many.

    Thank you.

    Don of Tallahassee

  135. Don: yes, your Voynich Lite collection does work well as a collection. But does that make it a model? No, not really. I’m hoping – really hoping – that good models will start appearing, because they should help us to answer difficult questions.

  136. Dear Mr. Pelling,

    “It does work well as a collection” of the Voynich Manuscript’s 505 most common words. What do you want to have a collection of, for the love of Pete?

    Are these “good models” you write of real or something you are only hoping for? Do you have any evidence such “good models” even exist for your arguments? If so, could you reveal one or more to me/us as evidence of your claims?

    Have these “good models” started “appearing” unto you yet? Or is that one of those “difficult questions” of which you write?

    Are you keeping these “good models” secret?

    It sounds to me like you are arguing on a hope and a prayer for some pie-in-the-sky “good model”, don’t you think?

    My (bad ?) model works for the real VMS words. I can’t yet tell if yours does or not. I would like something more than a wish sandwich from you with which to check your ideas and work.

    Please show me any alternate and reasonable way to account for the construction/deconstruction of Voynich words that can be consistently applied to the whole manuscript…or just the 505 most common words…if that is easier for you.

    Or show me where my ideas don’t work with the 505 words.

    I showed you mine. Your turn.

    Or you can blither, blather and bluster some more about your preconceived ideas about what the VMS must show you, as if you found those strictures chiseled into stone tablets or written in the Voynich Manuscript’s direction manual. That would be fun to read again, too.

    Please let me know if I seem to be arguing too hard or well for my ideas or if I offend you in any way. I try to argue fairly and based on the evidence, but well and with great vigor. I don’t want you upset with me.

    Separate topic – what was the thing you recently hinted you were looking at that I sent to you – the telescope thing? Did any more come of it?

    Thank you.

    Don of Tallahassee

  137. Don: the issue here is a matter of category, not of what I want or don’t want. “Blither, blather, and bluster” – I liked that, that was good. But here’s the deal:

    “Voynich Lite” is a dictionary generation template – it records (in a compact way) a way of generating a lot of Voynich words, but at the cost of containing a long list of pieces of them inside. It records Voynichese, but it does not explain them. As a recording of the dictionary, it is, as you say, fact-based: but it’s still not genuinely a model, in that it does not idealize out the underlying processes, or embody any kind of theory about the underlying processes. The first part of https://en.wikipedia.org/wiki/Statistical_model touches on a lot of the differences:

    A statistical model embodies a set of assumptions concerning the generation of the observed data, and similar data from a larger population. A model represents, often in considerably idealized form, the data-generating process. The model assumptions describe a set of probability distributions, some of which are assumed to adequately approximate the distribution from which a particular data set is sampled.

    A model is usually specified by mathematical equations that relate one or more random variables and possibly other non-random variables. As such, “a model is a formal representation of a theory” (Herman Adèr quoting Kenneth Bollen).[1]

    All statistical hypothesis tests and all statistical estimators are derived from statistical models. More generally, statistical models are part of the foundation of statistical inference.

    At the same time, we have a hundred or more statistical observations about Voynichese that puzzlingly point away from a single homogeneous one-size-fits-all rule for every word: for example, the difference in behaviour for labels, paragraph-initial letters, line-initial letters, line-final letters, “Neal sequences”, “titles”, A and B “languages”, word length between A & B pages, and so forth. And where are numbers, “and”, “the”, and countless other common words we would expect to see? Where are the consonants and vowels, even? Why is it that so many words in the dictionary differ by only a single letter?

    I think, broadly as John Tiltman concluded half a century ago, that this mass of statistical evidence points to a complicated set of processes going on in parallel “under the hood” of the cipher system: and there is a lot of modelling work that has yet to be done to make sense of this thing in its entirety. So far, I have ideas about only some of these processes (for example, that “4o” may well encode “lo” [i.e. ‘the’], and that the “aiiv” family may well encipher Arabic digits, etc). but I don’t mind admitting that constructing a model that encompasses all of them has so far been beyond my abilities, because it has plainly fallen beyond everyone else’s abilities too.

    So: far be it from me to stop you arguing with great vigor (which I actually applaud), but what you have done to date simply does not yet fall into the category of what I or others would call a theoretical model, in that while it records almost everything, it explains almost nothing. There’s a huge difference, and getting cross at me about it won’t bridge that gap.

    PS: the telescope thing is a paper I’m trying to write, but my time seems to get fractally filled with other stuff. 🙁

  138. boyfriend , Champollion,,. :-) on January 5, 2016 at 9:56 am said:

    Hi Nick.
    You are writing : Why is it that so many word in the dictionary differ by only a single letter ?

    You , as a great expert and scientist do not know why the manuscript a few, thos about 16 characters. From which consists of text and entire entry. Because I want to help and advise you of course , thos these 16 characters. Covers the whole alphabet.

    Write someplace those 16 characters, and I’ll show you the correct meaning. Otherwise, you will wander, many many, years.


  139. Dear Mr. Pelling,

    Yeah, I liked blither, blather and bluster, too. Nothing like a little alliteration for emphasis. I try to keep my writing interesting for the reader.

    As to the rest of your objections – I think you are still trying to impose your own conditions on the manuscript.

    I can find nowhere in the VMS that it says the manuscript needs to fulfill your expectations, wait for you to complete your modelling before it can be understood, or that it must (for some reason) answer your questions. On what folio did you find those requirements? Can you show me where it says that?

    BTW – how’s it going with your amorphous, diaphanous, vaporous and nebulous “good model” that you have trumpeted so blaringly? Has it “appeared” yet?

    I think it is surprising how few of your objections are based on anything except what you expect to find, don’t you?

    My ideas have been pre-rejected due to your superior knowledge and techniques that give you all the answers about why my ideas are incorrect (but not any of the answers to your questions). That is convenient for you, isn’t it?

    Just because my ideas seem to work – that’s no reason to consider them, if I understand your attitude.

    I wonder how all these things were revealed unto you? Was it a vision, a visitation or maybe handwriting appearing on the wall? Did a hand magically appear, write down your preconditions and then move on? If so, how nice for you.

    How do you come up with all these preconditions for a book you cannot read and that you know (or might have reason to suspect) has deliberate attempts to throw you off the track? How do you know all your preconditions must be met by the manuscript? Why, after a hundred years of no one being able to come up with any kind of “good model” that works, are you so quick to reject my (bad ?) model that does work?

    Here’s the really hard question – why haven’t you asked me about the possible answers to all your questions?

    I guess if your mind is already made up, my answers wouldn’t mean anything at all to you, huh?

    Even a dunderhead like me can understand that.

    Thank you.

    Don of Tallahassee

    PS – The part of this exchange I find most strange is that although you agree that my proposed way to deconstruct VMS words very consistently does generally seem to work – you just don’t appear to like it that it does. : )

    PPS – It was very nice of you to admit (way up in the second paragraph of this thread) that “Numerically, Don’s model works well;…”. You were the first to make that statement. I want to make sure you (and others) understand how much I appreciate your comment. Thank you very much.

  140. Don: you have gone from the facts to your (compact) collection, so of course it is a good fit. But if it is as good a model as you think it is, what does it tell you about Voynichese? Please feel free to tell me or anyone else, I’m all ears.

    The whole point of modelling is, by going beyond data collection, to reach towards the generative processes beneath and inside the data. From my experience, the right kind of question to be asking is: what causes Voynichese to look the way it does? The challenge is therefore one of identifying cause and effect.

    There is a difference between causality and consistency – what you have done is map the surface effects of Voynichese, but countless statistical studies over the last century have pointed out a whole realm of linkages and connections beneath the surface. if you can connect what you have done with those in ways that illuminate that which we can only indirectly see, then we’ll all be better off.

    But so far, all I can see in your Voynich Lite collection is a load of stuff that captures Voynichese as it appears on the surface, and skirts over all the interesting, tricky bit.

    The point of being a statistical researcher looking at Voynichese is that there is a mysterious sub-realm underlying all that surface – an oddly romantic inner life of data – that your comments and remarks often seem to somehow bypass. In cryptological terms, you are mapping the covertext: but the strong evidence is that there is a very large amount beneath that which is completely unmapped.

  141. Dear Mr. Pelling,

    I’m sorry. I haven’t seen any “mysterious sub-realm underlying all that surface – an oddly romantic inner life of data”.

    All I have to work with are the Voynich glyph words, They’re romantic and mysterious enough for me.

    I must ask if you are familiar with the rest of my ideas beyond the Voynich Lite/deconstruction sequence stuff?

    fumblydiddles.com does have a few other items pertinent to the Voynich Manuscript – several hundred pages of somewhat out of date stuff. Things like the proposed glyph/sound alphabet, the current assigned values for the number glyphs, some translation attempts, meanings for the codes in the Groups/Tables – that kind of stuff.

    If you’re not familiar with my site, you might want to skim it to see if anything grabs your attention or answers any of your questions. Voynich Lite was only the result of distilling a much larger panoply of ideas and results into one small, easily comprehensible example. Showing how half the words in the manuscript can be broken down using one system, one group of tables of code groups and one sequence is just a drop in the bucket compared to all the other stuff at my site.

    But, I must admit to being slightly biased on the subject of my site.

    I think “what causes Voynichese to look the way it does” is adhering to the principles for keeping secrets first set forward by Roger Bacon, shown at:


    Thank you.

    Don of Tallahassee

  142. boyfriend , Champollion,,. :-) on January 5, 2016 at 11:39 pm said:

    Dear Don.
    I looked up what you write. Well first, you have the wrong alphabet. Therefore, you will not be successful. Secondly, you must use substitution. Which has very clear rules.


Leave a Reply

Your email address will not be published. Required fields are marked *

Post navigation