The last few days have seen a huge flurry of Voynich-related Internet interest: a University of Arizona press release on radiocarbon testing the Voynich Manuscript’s vellum sparked a Discovery News item and a hundred or more slightly-edited reposts, with even Fox News getting in on the act yesterday. Sadly, though, it’s all the “same old same old”: you’d be forgiven for concluding that – the UofA’s whizzy new dating aside – there has been nothing new under the Voynich research sun since Mary D’Imperio’s (1976) “An Elegant Enigma”.

But actually, we know a huge amount now. For example, I think I can now prove to the satisfaction of 99% of historians that the Voynich Manuscript is indeed a ciphertext. Even though I first described the following in 2006 (The Curse of the Voynich, Chapter 10: “Secret Numbers”), nobody seems to have picked up on it as a basic historical proof.

So here’s my argument why it’s a ciphertext, one rock-solid step at a time.

(1) Medieval page references.

To make a medieval book, you fold a small double-width set of vellum or paper leaves around a single central line: this unbound-gathered-together nested set is called a gathering. Then, you pass off a set of gatherings to a binder, each gathering with its own sequential mark – typically ‘a’, ‘b’, ‘c’, etc – so that you can be sure the binder will bind them together in the right order. Once sewn and bound, the gatherings become known as quires. Within each quire, the individual folios (leaves) are typically numbered ‘i’, ‘ii’, ‘iii’, etc, while the two sides of each folio were referred to as recto (front) and verso (back), usually abbreviated to ‘r’ and ‘v’ respectively.

So, when medieval writers wanted to refer to individual pages, the ‘address’ of that page was its quire index (a/b/c/…), its folio number (i/ii/iii/…) and its side (r/v). For convenience, these three elements were usually fused together into a short string, i.e. the first few pages of the first (i.e. the a-th) quire would be referred to as air, aiv, aiir, aiiv, aiiir, aiiiv, etc. Everyone who worked with manuscripts knew these were medieval page references: this was the cultural norm.

(2) Voynichese appears to be full of medieval page references.

For someone circa 1450 looking at the Voynich Manuscript, one single feature of its mysterious text would have almost jumped off the page – the medieval page references, all apparently to the first (a-th) quire. For example, here you can see an “aiiv” group with an “air” group immediately below it:-

Whether you like it or not, there it is: the Voynich is utterly stuffed full of these medieval page references. However…

(3) These don’t function like medieval page references.

If you count these up, you discover that aiv/aiiv/aiiiv (1675/3742/106) occur more than eight times as often as air/aiir/aiiir (564/112/1). Simply put, the stats are wrong. Also, there appear to be no references to other quires, just to the a-th quire. 

(4) These are not medieval page references

So, if these things so closely resemble medieval page references (but aren’t), what are they? The famous WWII codebreaker Brigadier John Tiltman wondered whether these might actually some kind of obfuscated Roman numeral scheme, but the stats were wrong for that idea too. Hence they don’t match any obvious numerical or referential scheme.

(5) These appear to be integrated within the text

Even if we cannot directly read them, we can see that they recur throughout the text, and – as in the example shown above – that they are often integrated with other Voynichese letters into words. But… if they appear as consistent groups, are integrated within words, and resemble something that they clearly aren’t (i.e. medieval page references), then the only rational historical inference is that…

(6) These are cipher shapes.

QED. But… if even one element of the writing scheme is written in cipher, then the conclusion has to be that…

(7) The Voynich Manuscript is a ciphertext.

Again, QED. However, even though medieval page references were known circa 1450, their heyday had passed: they were far more part of the medieval world of the monastery scriptorium than the emerging Renaissance world. Hence the presence of fake medieval page references in its ciphertext inexorably leads (I believe) to a further curious conclusion:

(8) Voynichese was constructed to resemble a medieval document in a fake archaic language

I contend that this is the only logical conclusion consistent with the radiocarbon dating: an early Renaissance mind looking back at the Middle Ages, cunningly appropriating medieval textual tropes for his/her ciphertext’s alphabet.

Now, please tell me – which part of this argument do you disagree with?

116 thoughts on “Why the Voynich Manuscript is a ciphertext…

  1. Those two words.Out of context.!!
    (oNam = known)
    (Man R = Men R =mercurial (mercury).

  2. “an early Renaissance mind looking back at the Middle Ages, cunningly appropriating medieval textual tropes for his/her ciphertext’s alphabet.

    Now, please tell me – which part of this argument do you disagree with?”

    None: I’m all for the idea this book has many “cunning appropriations”… in fact, that it is almost composed of them. Rich.

  3. *** Hola Nick: Parece una interesante teoría y coincido. Hay pruebas en otros manuscritos.

    *** Hi Nick: Looks like an interesting theory and I agree. There is evidence in other manuscripts.

  4. Vytautas on February 17, 2011 at 10:33 am said:

    In the set of these “cunning appropriations” are botanic (plants), astrology (Zodiac), possible Geomantic (elements in counts 16 and 12), astronomy (if only stars are astronomical) and at last but not least – medieval page references. I think Mr. Gordon Rugg will have more peoples at his side in near future 🙂 because of this suffle of ancient things. But who can propose these things have really these meanings ? Page references may represent state changes in cipher as we talked here some time before…

  5. Artur Sixto on February 17, 2011 at 11:02 am said:

    Hi Nick,

    I am new to this (VM, cipher, etc) but do like puzzles.

    1) Not being knowledgeable with medieval page references, I fail to see why they surprise you in a manuscript dating back to an age when printing was just about to be born.

    2) I guess the author might refer to pages in other manuscripts than his own.

    3) I guess word repeats might relate to ritualistic, sort of incantatory repetitions, rather than linguistic reasons.

    4) I guess the author could have a professional link with bookmaking.

    5) I guess the person could be Gutenberg himself (who apparently delved in secrets around 1450): http://en.wikipedia.org/wiki/Johannes_Gutenberg

    6) I guess his invention of movable type may have been born out of a search for a modern cipher tool. It does look very apt to scramble letters and information in a modern, mechanical way.

    7) If Gutenberg was smart enough to invent printing he might have been smart enough to author the VM. Who knows, maybe its text describes the design of a printing press and recurrently expresses Gutenberg’s joy at his own clever invention, like: God! God! God!

    8) Woodblock printing antedates Gutenberg’s and word of it might have inspired Gutenberg. http://en.wikipedia.org/wiki/Woodblock_printing

    Now please tell me if so many guesses of mine are fun or not? (Maybe they were already made by others and ruled out.)

    Artur

  6. Vytautas & Rich: I don’t think there’s any grounds for watering down what is quite a strong argument. Why would a hoax present such a well-defined statistical profile specifically for these medieval page reference-like features that is without much doubt quite unlike the statistical profile of page references in real documents? The notion that this is a simulacrum of a medieval herbal isn’t supported by the statistics, even though the medieval page reference pattern has been appropriated. That is, the author duplicated the shapes but not the statistics.

    Vytautas: yes, these could represent state changes in the cipher (as could any symbol) – but right now, I’m more concerned about demonstrating clearly that Voynichese is indeed a cipher. That alone would be a big step forward in the overall discourse.

  7. *** Nick: Puedes demostrarlo con la ley de Zipf, o no es prueba suficiente? Si no es así, que hace falta para demostrarlo?

    *** Nick: You can demonstrate this with the Zipf law, or is not proof enough? If not, we need to prove?

  8. *** Artur: Gutenberg? Es posible. Pero se requiere algo más para realizar esta apuesta.

    *** Artur: Gutenberg? Maybe. Why?

  9. Artur: so many guesses in a row – you’re practically a Voynichologist already! 😉 Of course, the difficult bit with guesses is putting the work in to find out how the evidence supports (or contradicts) them: but don’t worry, I’ll do that bit for you. 🙂

    It turns out that, as a person, Gutenberg is a surprisingly good match: he was a secretive, literate goldsmith and knew the Master of the Playing Cards (who, crucially, used parallel hatching). Gutenberg’s dates are pretty close to the VMs’ vellum’s radiocarbon dates, too: and Rene Zandbergen has long proposed that the author of the VMs may well have been German. On the downside, this is exactly the kind of join-the-dots “Big Man” History that got hopeful Victorian historians into such trouble, so we should be extremely careful! 🙁

    Then again, the VMs’ author could just as well be Laurens Janszoon Coster (from whom Gutenberg may possibly have cribbed the idea of printing). It’s hard to tell without any kind of signature. 😉

  10. Artur Sixto on February 17, 2011 at 11:54 am said:

    Hi Nick,

    I am really happy to see that my ideas could be of use (I know about the VM since Monday only).

    I will be even more happy to let you do the “bit” of work of seeing whether all these guesses are really worth.

    I am baffled though by who could have written the months in Catalan or Occitan. (I am Catalan.)

    Kind regards,
    Artur 🙂

  11. Artur: the zodiac month names appear to have been added by a quite different (and probably later) hand, I’d guess later in the 15th century. The spelling of “augst” may indicate that this person was exposed to both Occitan and German, in (say) Savoy or Switzerland – again, it’s hard to say.

  12. Artur Sixto on February 17, 2011 at 1:03 pm said:

    Hi Nick,

    To me the months seem to correspond slightly better to Catalan than Occitan. June for instance, spelled with “ou” corresponds to Catalan pronunciation, in French writing. “ny” would be Catalan relative to Occitan “nh” or French/Italian “gn”. So the person might have ties with the North of Catalonia (and could have a French influence).

    I have just seen that the person who took Coster’s secret to Mainz was named Fust but usually spelled as Faust. Faust is a name in Catalan, spelled exactly like that (Faust). 30 people born in Catalonia bear it as surname. http://www.ine.es/fapel/FAPEL.INICIO

    Fust is also a Catalan word (and may have been a surname) meaning wood. Fuster is a very common surname in Catalonia, meaning carpenter.

    On top of that, first documented printer in Haarlem (and Coster successor) was Jacob Bellart. Today more than 300 people born in the North of Catalonia (Spanish side of the border) bear this surname. He may have been from a Jewish family that fled the progroms that erupted in 1391.

    Interestingly, many Jews in Catalonia spoke Catalanic, a Catalan dialect close to Shuadit, i.e. Judaeo-Provençal (i.e. Judaeo-Occitan).

    I am wondering whether the VM might have been from Coster and taken by Faust. Faust or someone else of Catalan descent in the Netherlands or in Mainz could have added the months.

    🙂 Artur

  13. Artur: only once somebody does some proper high resolution multispectral scans on the Voynich’s marginalia (specifically f17r, f66r and f116v) will we stand a reasonable chance of isolating the original author’s language. And if that proves to be Catalanic, well… 🙂

  14. Artur: Crec que tens tota la raó, però si et fixes en els escrits de llengua vernacla, en cap cas han escrit sobre ells en voynichés, fins i tot en els textos que s’han pretès esborrar. Això dóna a suposar que no són de l’autor original.

    És possible que el MS-408 fora propietat d’algun català, possiblement a Nàpols, ja que en alguns textos esmentats apareix alguna paraula que és més pròpia del italià que de l’occità. 🙂

  15. Sergi & Artur: I was going to introduce you two anyway, but Sergi got there first. 🙂

  16. Artur Sixto on February 17, 2011 at 3:17 pm said:

    Hola Sergi i Nick / Hi Sergi and Nick:

    Encantat de conèixer aquí un compatriota català / Nice to get to know a fellow Catalan here.

    Nick, how reliable is the 14C dating? I’ve read the VM was sampled just where reader fingers would have contaminated it with grease. This would imply a dating shift towards the present, and the VM could then be older, like XIVth century more than XVth. In such case an excellent candidate to have authored the book would be Arnau de Vilanova: http://en.wikipedia.org/wiki/Arnaldus_de_Villa_Nova

    The guy was not only Catalan in all likelyhood, but above all a most authoritative alchemist and physyician at European level. He may therefore have had quite a few things to be kept secret. Not just in Alchemy, but maybe even more in Medicine. He may have had more reasons to fear prosecution because of medical knowledge (particularly if obtained by dissection). Remember the Spaniard Miguel Servet was burnt at the stake much later (in Switzerland) just for proposing blood circulation, against some kind of religious dogma.

    Now, what do we see in the funny drawings of the VM? We see “pipe” sections connecting different parts that look organic. It could therefore well be that the drawing itself is kind of cipher, like a map of blood vessels in disguise, under the looks of simple pipes.

    To me, the plants look mostly invented, maybe to conceal the fact that they constitute a catalog of mandrake roots. As you probably know, mandrake and mandrake roots enjoyed a pretty nice popularity at the time, particularly among Arnau de Vilanova’s colleagues.

    Therefore, I think that not only the text is likely cipher, but also the drawings, which in part could convey information in the middle of much distracting fantasy. Why put so much effort otherwise, in so many pages of text and illustrations if it were all a joke?

    Artur 🙂

  17. Artur: sadly, we’re still waiting for a proper paper discussing the sampling and processing methodologies used – the scientific machinery behind the headline dates. However, it does seem likely to me that they aggressively stripped the vellum back to (basically) gelatin: they also took four samples from different pages in different sections, and (I expect) avoided sampling from any sections that presented any obvious sign of contamination or stains.

    I’m not sure that Arnau de Vilanova is going to work as a realistic Voynich candidate: the radiocarbon dates (1404-1438) fall pretty much a century after his dates (1235–1311), so probably out of range. But an interesting guy all the same. 🙂

  18. Artur Sixto on February 17, 2011 at 3:48 pm said:

    Nick: I’ve seen big enough 14C discrepancies between labs in other cases, and the issue of vellum margin contamination by fingers doesn’t help.

    I don’t know who the author/s is/are going to be, but I wouldn’t rule out A. de Vilanova and his contemporaries as candidates.

    BTW, the tubs where women are supposedly bathing could have to do with winemaking. Who said it’s got to be water? (And Arnau was an expert in wines.) OK, maybe not with winemaking but maybe with wine and health? Skin health? “Wellness” of some kind? 🙂 Or baptism if you prefer. I wouldn’t argue without a clue of what the text is all about.

    a

  19. Artur: I completely agree, but the “finger oil” contamination issue might be something their sampling and processing methodology was designed to help mitigate. Until they (finally) get around to writing up exactly what they did, we simply can’t assess how reliable their headline conclusions are. 🙁

  20. Artur Sixto on February 17, 2011 at 5:53 pm said:

    Nick: a few more things before I go.

    1) Text could have nothing to do with the drawings.

    2) But text seems too long to have anything to do with politics or spies. When I read about Dee having been influential in English politics at high level, I thought he might have travelled to central Europe as a top level spy/diplomat. I doubt though, he should need to carry such a lengthy written information. The fact that vellum is at least a couple of centuries older doesn’t rule him out completely though, because an old blank parchment could have been used.

    3) Parchment leftovers must have circulated for quite a long while. The VM parchment could be theoretically from the XVth century, the drawings from the XVIth century, and text from the XVIIth centuries. Vellum was expensive and books a luxury. A scriptorium could not deliver a book with damaged vellum. Damaged vellum could be of little use but still be kept and further deteriorate. Then someone could use it to draw silly things (someone bored or intoxicated). Then someone could use it to write cipher between the drawings (or gobbledygook for selling to an emperor with Kelley’s sort of cheek).

    4) Arnau de Vilanova could also have been involved in some high level secret mission. When he went to teach Medicine in Montpellier the region was under the House of Barcelona. From there he later went to Paris. Arnau de Vilanova was advocating Raimon Llull’s Rex Bellator project, which conflicted with the French crown interests. Still, same thing, I can’t make sense of such a lengthy text. Unless it was a way to record secret information over some period of time, to be sent back home.

    5) Writing can be fluid without speed. Proof is: we know a quill was used (which isn’t fast on parchment) and it does “look fluid”. Someone could proceed alone with relative calm and fluidity, or else with another person dictating (either way it’s just a matter of memorizing short strings of characters one after another).

    6) Real fluidity could otherwise point to a simple cipher which should have been solved long ago. Unless you tell me criptographers couldn’t necessarily breake a code simple enough for a learned person to mix latin, greek, arabic, characters etc, plus say, invert syllables at the same time, as he wrote.

    7) Interestingly, Arnau de Vilanova has various connections with the older world of say, Ptolemy (reg. the naked women style). In Arnau’s work “Hermetis Trismegisti Phoenicum Aegyptiorum…” is included a very nice Tabula Hermeticae Phoenicio using Phoenician characters some of which are identical or very similar to particular Voynichese characters by simple rotation or rounding of angles.

    8) Beneventan script seems to have been still in use as late the XVIth century. This and other related scripts could have been used more or less naturally / fakedly.

    That’s all folks! 🙂
    a

  21. Artur Sixto on February 18, 2011 at 9:44 am said:

    Hi All,

    Sorry, “Hermes Trismegisti…” quoted Arnau but wasn’t from him.
    http://www.slub-dresden.de/index.php?id=5363&tx_dlf%5Bid%5D=8884

    But it does include the Emerald Tablet (or Smaragdine Table) in Phoenician script. This script you can see in the first page of the above-mentioned book, with the link provided. It has a special appeal for several reasons:

    1) some characters close to Voynichese
    2) a seminal alphabet in the ancient world (classical prestige)
    3) consonantic (like Hebrew)
    4) remote (though likely alive in Lybia up to XIth century)
    5) maybe undecipherable then, but known of at least, by medieval savants from the Western Mediterranean (particuliarly if familiar with the islands of Sicily and Sardinia).
    6) mithology, cults, literature, etc. traceable to Phoenicians as ultimate source of knowledge. Please confer the Praeparatio Evangelica and Sanchuniathon’s oldest sacred lore, based on secret writings.
    7) an old amalgamating culture of seafarers.
    8) settlements overlapping with later Ptolemaic world.
    9) smaller coincidental things, like Phoenician ships being called galloi (i.e. tubs) in Greek, because of their shape.

    In summary Phoenician might hold clues about Voynichese script. Is it possible that a learned person could have concocted and used of a mixture of characters from different scripts? Some hybrid alphabet to write directly or with small added tricks? This would be a natural language. It could be any language (Latin, Hebrew, Catalan, Italian, etc.) transliterated into this hybrid alphabet. Like using a phonetic script and maybe inverting syllables or something else. A learned person with good command of ancient languages could have played around with them.

    At this point I am thinking of another candidate author than Gutenberg. BTW, if nodody had previously thought of Gutenberg it must be because of his name being so intimately associated with printed books instead of handwritten. Fact is though, that most or all of his personal stuff Gutenberg would handwrite instead of print.

    Anyway, the other candidate I’ve thought of, actually much better than Gutenberg, is Pico della Mirandola. For a lot of reasons! Knowledge of languages, incredible memory, interests in Hermetism, Kabbalah and everything, papal accusations of heresy and scandalous ideas.

    Hope this inspires someone! 🙂
    Artur

  22. * Hola Artur: totes aquestes teories estan molt bé i tenen certa lògica.

    Jo intento seguir les següents pautes:
    – Miro aquest blog o per Internet les diverses teories que hi ha, per si alguna coincideix amb les meves o aprenc alguna cosa nova.
    – Després busco proves o almenys alguna cosa que pugui insinuar.

    Tinc d’advertir que en general, les teories que no hi hagi Jon Dee o el Rei de Bohèmia, no són molt populars. No entenen que a l’Europa del segle XII-XVI, la Corona d’Aragó era una potència a la Mediterrània, tant d’extensió, com a militar, econòmica, cultural …

    Això és gràcies a la censura durant segles del centralisme espanyol, que no reconeixen la importància de cada regne. Ramon Llull (segle XII) és un exemple, un dels més grans científics del món, reconegut com el pare de la informàtica, Espanya no se li presta cap atenció, els millors estudis d’aquest científic s’han realitzat a Alemanya.

    * Hi Arthur, all these theories are fine and have a certain logic.

    I try to follow these guidelines:
    – I look at this blog or the Internet there are several theories, if any of my matches and I learn something new.
    – then look for evidence or at least something that might insinuate.

    I notice that in general, theories that there is Jon Dee and King of Bohemia, are not very popular. They do not understand that the twelfth-sixteenth century Europe, the Crown of Aragon was a power in the Mediterranean, so long, as the military, economic, cultural …

    This is censorship through centuries of Spanish centralism, which does not recognize the importance of each kingdom. Ramon Llull (twelfth century) is an example, one of the greatest scientists of the world, recognized as the father of computing, Spain will not be paid any attention, the best scientific studies that have been made in Germany.

  23. Artur Sixto on February 18, 2011 at 10:36 am said:

    Just thought of more things…

    Some drawing features are reminiscent of XIVth century. Crossbow man for instance. Not just his weapon (I don’t know about them) but his clothes and the actual drawing style. I don’t trust the 14C results anounced. Firstly because of the delay in publishing. Secondly because this manuscript is probably, among all incunabula of value, the most groped of them all, with the most cheesy edges.

    Nothing special against UA but one shouldn’t trust much 14C results until knowing the Material and Methods and if possible, see results confirmed by a second lab.

    One shouldn’t be surprised if the vellum turned out to be from as early as the XIIIth century. Can you imagine all the possibilities this could open? For instance, in the mid XIIIth century, 3-4 cathars escaped besieged Montsegur with something of great value to them, possibly some books. One can dream the book has to do with Cathars, Templars, Hopsitallers, whatever! We can’t close any door, really. It’s fantastic. This VMs is something really great.

    Nick, hope you’ll send me your next book on the subject! 🙂
    a

  24. Artur: the zodiac medallion drawings (i.e. the pictures in the middle) have always been an awkward thing to deal with: whereas all the nymphs and baths in Q13 (the water quire) have a 1400-1500 early Italian Renaissance feel to them, the zodiac medallions come across as both Germanic and medieval, say 1300-1400 or so. Hence for many years, the received wisdom in Voynich research land has been that the pictures on these zodiac medallions were perhaps copied from an earlier German woodcut almanack. But this is just an hypothesis! If the Voynich’s author was German… 🙂

    Now, I have to point out that I don’t currently believe that the Voynich Manuscript was written in Germany – influences are rarely that simple. But who knows, eh? 🙂

  25. >Which part of this argument do you disagree with?
    I honestly have a problem with step 2. Are you sure that these two strings contain “aiiv” and “air”? Are you sure that all these strings you mention are page references? Of course, I see a certain similarity. However, similarities happen. I also see similarities between Nostradamus’ verses and actual events, but I don’t believe that Nostradamus forsaw the future.

  26. Artur Sixto on February 18, 2011 at 5:15 pm said:

    Nick, a closer look at zodiac medallions strongly suggests the VMs cannot be a hoax. Reason: VMs seems put together by someone who rescued remnants of various similar books with variation in details, or at least that someone was concerned to save information that would have been more obvious in the past, which points to an agonising cult. Either a prosecuted cult or some occult knowledge in decline, hence the secrecy.

    I counted three months of “April”, like someone had trouble assigning months to rescued pages, and counted FOUR rams: all of them munching the same way, 3 of them undoubtedly Aries (penis included) while the fourth could be Capricorn (a goat instead of a ram). Let’s compare these medallions.

    1) In f70v1 it’s “April” (abril) and Aries, with some women dressed and other naked, all inside tubs.
    2) In f71r it’s “April” (abril) and Aries or Capricorn, with women dressed inside tubs and a few dressed men inside tubs save one, naked or unfinished.
    3) In f72v3,v2 it’s “April” (abril) and Virgo, and full of naked entire women.
    4) In 71v it’s “May” (maig) and Aries, with women dressed inside tubs.
    5) In 72r1 it’s “May” (maig) and Aries, with naked women some standing entire others inside tubs.

    Finally I can’t resist to mention “November” because it shows a nice dragon that doesn’t match any zodiac sign, dragons in Catalonia have a milenary tradition we commemorate in April, and it’s got more naked women than all other medallions.

    I don’t know if all the Voynichese comes in a single handwriting. Maybe yes, and drawings also seem to come from a single person (the zodiac medallions at least). But then, the same author could have made several books, and their remnants be put together quite later.

    Anyway, seeing the mess, a hoax seems utterly unlikely whether the text says something or not. Don’t you agree?

    I think it’s genuine old stuff, for all the bizarre later-looking jars or optical devices in some pages.

    a 🙂

  27. Klaus: To be precise, I’m sure that for someone looking at Voynichese circa 1450, the “aiiv” and “aiir” strings overwhelmingly resembled medieval page references, yet I’m also sure that these same strings do not actually function as medieval page references.

    Comparing this with Nostradamus’ quatrains is not a great counterargument: the mirroring exists in the conscious design of the aiiv and aiir shapes, which appear on pretty much every page in the manuscript. That is, the similarity I’m pointing out is not a matter of after-the-event selective interpretation, it’s a matter of historical understanding – the more you understand medieval codicology, the more certain you become.

  28. Artur: optical devices? I really don’t think so!

  29. Artur Sixto on February 18, 2011 at 6:55 pm said:

    Has anybody physically checked whether the VMs is a collection of book parts that someone put together? I suppose yes, so is it the case or not that different parts could physically come from different books?

    If the answer is yes, was this taken into account when performing statistical analyses? Could we refine them analysing separately those parts that could come from various books?

    More questions. Has anybody compared VMs stats with stats of different sorts of text? Section by section? If parts of the VMs contain prayers or incantations, they could be more repetitive. Even according to original language and tradition. See for example how muslims tend to repeat conventional expressions attaching to Allah every time Allah is mentioned.

    A section dealing with prayers could have slightly or markedly different statistical patterns in word characteristics from a section about recipes, plant descriptions, cosmological or mithological explanations, etc.

    BTW Nick, I didn’t say it’s optical devices. I said it’s funny they look like either jars (could be smaller recipients also) or optical devices from more recent times. I don’t know what they are. It could be oriental, byzantine, or whatever.

    The drawings could be a nonsensical facade behind which the VMs could even conceal muslim writings more than Hebrew, Hermetic, Astrological or Alchemical beliefs. In the Iberian peninsula muslims may have been more discriminated against than jews, having to convert or leave. They were not supposed to draw naked women, but despair to save writings could lead to unexpected decisions. The pipes and women could relate to muslim paradise, with plenty of houri for everyone plus honey and milk flowing all around.

    The VMs could have roots in Northern Africa, Near East, Byzantium, Persia… It could be inconvenient text from whatever origin (like ancient Greece) processed by any hands, anywhere. It could be Aristophanes or whatever. The drawings could have been made by some German in Jerusalem or Constantinople, or by an Arab in London. Just like today. 🙂

    Artur

  30. Artur: whenever you find yourself slipping into the past conditional, give yourself a good slap and tell yourself to concentrate on what it is. A pinch of “is” beats a ton of “could haves” every time. You know it makes sense. 🙂

  31. >I’m sure that for someone looking at Voynichese
    >circa 1450, the “aiiv” and “aiir” strings
    >overwhelmingly resembled medieval page references
    It is difficult for me to check, if this is true. Do you think you can hand in an article on your theory at Cryptologia? It would be interesting to see, if the peer reviewers accept it.

  32. Klaus: I’ll see what I can do… but it’s more of an historical argument than a cryptological one.

  33. Nick. Artur has in some truth.
    The drawings could be a nonsensical fasade.
    Drawing = symbol (alchemy)

  34. *** Artur y Josef: Los dibujos que aparecen en el Voynich, tienen sentido, no están codificados. Hay que observar ilustraciones y leer manuscritos de alquimia de la misma época.

    No pueden ser falsos; mantienen los ordenes establecidos y estandarizados, se encuentran las ilustraciones ordenadas de forma jerárquica.

    *** Artur y Josef: The pictures displayed on the Voynich, make sense, are not encrypted. You have to view pictures and read manuscripts of alchemy in the same period.

    Can not be false, maintain the order established and standardized, illustrations are arranged hierarchically

  35. Sergi.
    I’m right. Look at my site.
    Microscopes. Women. Plants.
    (iluze – deziluze)

    Hola.

  36. Rene Zandbergen on February 21, 2011 at 11:40 am said:

    Hi Nick,

    reading through your logic, when I arrive at point (8), my conclusion would be: it’s a fake meant to look like a medieval MS…. I don’t see anything pointing to cipher.

    I also have a more practical problem.

    Beside:
    air, aiir, aiiir
    aiv, aiiv, aiiiv

    there are also:
    ar
    av
    al, ail, aiil, aiiil
    am, aim, aiim, aiiim

    This shows that the folio reference hypothesis only stands when one ignores part of the evidence. In particular, ‘ar’ and ‘al’ are very frequent.

    Furthermore, it does not make a lot of sense that page references only occur to quire “a” and folios 1-3…

    Cheers, Rene

  37. Rene: you’re correct… only if you ignore the stats. If the text was supposed to replicate / imitate actual medieval documents, then you would predict that the aiir and aiiv families ought to appear roughly as often as each other, yet the latter appears 8x more often. Why would a fastidious faker introduce such a skewed distribution to spoil his/her otherwise delightful illusion? This is why it can only be a cipher: the VMs combines the appearance of a medieval document with an entirely different statistical profile.

  38. Rene Zandbergen on February 21, 2011 at 1:36 pm said:

    Hi Nick,

    two different points:

    – I’d be quite interested to see examples of such folio references in other documents.

    – to those who have doubts about the C-14 results: some healthy skepticism is never bad, but one should not forget that this was not done by amateurs. Anything that we may come up with is something that they learned about in first grade, so to speak….
    And indeed, if their cleaning procedures could not eliminate all newer carbon, the actual age of the MS will only be older.
    How many % of early 20th C carbon would be required to move the 1420 date by a century or more, I leave as an exercise to the reader 🙂

  39. Rene: I’m already on the case as to your first point. 🙂 But as to your second, I’d also note – from the graph in the documentary – that the three later datapoints were clustered in the 1420s, while the earliest stood at about 1400, which would move the average of the four backwards a bit if it was some kind of outlier (do you know which folio that was from?) I’m not doubting the UofA people’s scientific ability, I just want to be sure that their historic reasoning is no less rigorous… as you are aware, plenty of clever people have made radiocarbon-related gaffes in the past. 🙁

  40. Artur Sixto on February 21, 2011 at 3:13 pm said:

    Hi, Nick and Rene.

    Reg. 14C people at UA surely know what they do, only WE don’t know what they did, and even after we know it’s good to remember that differences between labs do happen.

    Reg. the hypothesis of a fake I couldn’t find the McCrone Institute results online but they are supposed to rule out such hypothesis. Could someone please help me find those results as well as Mary D’Imperio’s work provided they exist online, or else tell me they are not available on internet? Thank you! 🙂

  41. René,
    Author manuscript did not use to write letters of the alphabet -( i,,,y,,). Instead, use the character – (q,,,a,,).Gematria = 1= a,,,i,,,j,,,q,,,y,,,./ And the entire manuscript/. I give the site the Voynich alphabet.
    Hi,,,Hola.

  42. Artur Sixto on February 21, 2011 at 4:25 pm said:

    Nick, I have some objections or doubts concerning your 8-point explanation:
    1) Those characters we describe as aiv, air, etc., we aren’t actually sure what characters they are. (We just call them a, i, v and r.)
    2) If we could be sure it’s actually aiv, air, etc., then we don’t know what they mean.
    3) It would seem inconsistent to encrypt over a hundred pages but leave characters a, i, v and r meaning exactly that (a, i, v and r) referring to pages of interest.
    4) “Words” you refer to, integrating or not “page references”, aren’t necessarily words. They are strings, bits of a sequence that we don’t know whether it’s constituted by words, and if it is, we don’t know whether spaces separate words or follow some rule to make deciphering harder (ora rej usta rbit rary).
    5) If text is cipher you shouldn’t be surprised that some page references may be 8x more frequent than others, as the author may have needed for whatever reasons to point 8x more to those pages.
    6) I can’t see on what basis you make the deduction that these are actually not page references. To me they could be, particularly in any language where given “letters” can also be used for numerals, especially when their frequency is at odds with other combinations of the same characters.
    7) Having said that, these characters (a, i, v, r) may have nothing to do with page references or numbers and their higher frequency could just be due to “language” rules. We know the “text” presents some rules regarding character location within “words” and between “words”
    8) I can’t see on what basis their not being page references is contributing so strongly to your conviction the text is cipher.
    9) If they were page references, reason why verso pages are cited 8x more often than recto pages would be a reason we obvioulsy can’t know in advance of deciphering.
    10) A Renaissance cipher would be right to try to appear as older script to help remove suspicions from the carrier of the cipher if they got caught (this is why Dee/Kelley for instance, could have been spies taking the book or writing the book in continental Europe).
    11) But nº 10 only applies if the VMs is proven to be from the Renaissance period. To me the possibility of an older manuscript/cipher cannot yet be ruled out, for all the UofA anouncements in the media.

    Cheers,
    a 🙂

  43. Artur: It all comes down to your #8: if they look like page references but don’t function like page references, then they are something else masquerading as page references – ergo, in cipher.

  44. Nick, nedělej tajnosti .
    Špatně to publikuješ.
    Vynechal si znak písmene ,,a,,.
    To se nedělá. Já se ti snažím pomoc a ty …/-

  45. Artur Sixto on February 21, 2011 at 5:33 pm said:

    Nick, the problem is YOU have decided that (1) they are not but (2) they masquerade as being so.

    What would be the reason to masquerade anything? Why masquerade they point to pages and thus give a clue it is cipher when the whole issue would be to conceal it is cipher (and remove suspicions from the carriers)?

    I am not knowledgeable about late Middle Age / Renaissance page numbering, but I can apply logic to things you say if it helps. If you think the text is Renaissance cipher, and if you think page references are of an older medieval type, then these are reasons to believe we are looking at actual page references instead of the contrary.

    Why? Two reasons. (1) Nobody seizing or subreptitiously looking at the book in the Renaissance would recognize and pay attention to aiv, etc as page references, in a book full of them, their being medieval and out of use. (2) If someone seized or peeped into the book, they wouldn’t be smart enough to do the stats proving that aiv, aiiv and aiiiv are 8x more frequent than air, aiir and aiiir, and could never use this as proof of spying against the book carrier.

    Conclusion: it may well be that aiv, aiiv and aiiiv point to actual pages. It is the only explanation I know of so far (together with my #7 above) that can account for their stats. If it happened to be the case then it would be a hint that we may indeed be looking at Renaissance or later ciphering, because a medieval encryption pretending to pass as an exotic language should have wanted to avoid obvious clues to page referral.

    Cheers,
    a 🙂

  46. Artur: that’s not a problem, that’s just how it is – if it looks like a particular thing but doesn’t function like that thing, I contend that it’s a second thing disguised as that first thing – which is another way of saying that it’s in cipher. Oh, and it can’t be 16th century or later (independently of the as-yet-unpublished microscopy patina studies), because it has 15th century number forms added for the quire numbering – John Manly knew this in 1931.

  47. Artur Sixto on February 21, 2011 at 6:34 pm said:

    Nick, you are the expert but I’m afraid you are being inconsistent here and I’m only trying to help.

    1) If it looks like a particular thing but isn’t, this doesn’t prove someone intended things to look that way as it could just happen by chance, like according to #7 or anything else. Take another explanation: aiv,aiiv and aiiiv may not point to pages or numbering and still need for whatever reason to be repeated 8x more often than air, aiir and aiiir. Suppose v/r encode sex, aiv, etc. referring to King, Marquis and Duke (or father, son and grandson) while air, aiir and aiiir would refer to Queen, Marquise and Duchess (or mother, daughter and grandaughter). The message may need to refer 8x more to the King (or father) than to the Queen (or mother) right? To roots instead of leaves. To Moon phases instead of seasons.There you have examples where it has nothing to do with pages or even numbering.

    2) Where is proof that aiv, aiiv and aiiv do not function as page references? (I gave reasons as to how they could actually be.)

    3) How do you know that 15th century number forms added to quire nbers. were actually added in the 15th century instead of later, or that encrypted text was written in the same period the book was bound?

    4) You think (rightly) that someone in the 15th century might want to masquerade as 14th or earlier, but reject that someone in the 16th or 17th century might masquerade as 15th?

  48. Artur: 1. Errrm… that’s enciphering (or, more accurately, encoding)
    2. They do not function in the same way that page references do, hence they are without any real doubt not page references. Also: repeated sequences of the same page number would be extremely improbable, yet we see this in Voynichese.
    3. The quire numbers were added in a 15th century hand, using 15th century number forms, and using an extremely rare Roman-Arabic hybrid numbering system that would match the 15th century time frame.
    4. Yes, unless they were a 16th century forger of such extraordinary sophistication that they would manage to fool 21st century experts. And anyway, it would require a 16th century forger to have constructed a 15th century document that in itself had been constructed to look like a 14th century document. That’s just too many implausibilities in a row for me, but perhaps your historical sensibilities are quite different from mine. 🙂

  49. Artur Sixto on February 21, 2011 at 7:39 pm said:

    Nick,
    2) May I ask again where is proof they do not function as page numbers? Also: why would it be unlikely to refer more times (or even exclusively) to given pages if need be, if information were more relevant or abundant in said pages?
    3) I am not following you. You say all quire/page numbers point to 15th century but said earlier the quire code is medieval instead of Renaissance. Anyway, if all quire/page number characteristics are consistent with the 15th century then what? It only proves they are not from earlier times.
    4) Better have 16th century people fool us than 15th or earlier, no? Anyway, I wouldn’t take offence if a 15th century criptographer made us believe he was from the 14th century. Neither would I take offence if a 16th century criptographer made us believe he was from the 15th century.
    5) It’s much the contrary. The more astute the criptographer was (if it’s cipher as we tend to believe) the more thankful we must be for giving us so much fun. But chances are the criptographer was not so devilishly astute, but just hit a nice solution we haven’t been able to break so far. And let’s face it: some codes are simply impossible to break unless you have the key.

    Hopefully we can break the code and it will be fantastic to understand it and know what it hid. But it could never happen and still not be any forgery nor any document devoid of hidden information. 🙂

  50. Artur Sixto on February 21, 2011 at 7:59 pm said:

    BTW, could somebody please tell me where can I download or read at least Mary D’Imperio’s work? Any advice as to how can I obtain it? No 14C publication, no microscopy patina study publiocation, no access to key research of 1976… Is it all a 21st century joke or do we really want to solve the mystery? 🙂 🙂 🙂
    Nick, would you mind using my email address to get in touch with me? I would be grateful if you did as I need to say something to you. Thx! 🙂

  51. Artur: 2. it depends on what kind of proof you want. I’ve given you as strong a proof as I think is reasonable.
    3. It proves that the Voynich was not made in the 16th century, quite independently from the radiocarbon dating.
    4. I’m not anti-16th or anti-17th century people, it’s just that the idea of a 16th century forger producing a work that looks like a 15th century work imitating a 14th century work complete with fake 15th century marginalia seems a bit too “airport novella” for me.
    5. I basically agree – I just wish that people would accept that it is (just as seems) a cipher, so that we could get on with trying to solve it, and not having to justify every methodological quarter-step.

  52. Rene Zandbergen on February 22, 2011 at 1:24 pm said:

    To #41 (Nick) and #42 (Artur):

    The averaging of the four samples was done in the ‘percentage C14 space’. Remember that a gaussian distribution there maps to an arbitrarily weird distribution in the ‘age’ space.
    The ‘oldest’ sample was the one shown in the documentary, of the astro foldout. This one had two peaks in the age distribution, the others had a much more skewed distribution with the more recent peak predominating. After averaging the four samples (and thereby reducing the standard deviation), only one peak was left (if you stick to 95% probability).

    The samples were chosen to maximise the possibility of finding different creation dates. After all, the MS could have been written over an extended time. This means: one folio in ‘A’ language, one in ‘B’ language, one of the thicker variety of vellum found in the herbal section. And then one foldout folio. The probability curves (in percentage space) fully overlapped. Of course, this still completely hides a possible 15-year creation time span as the individual measurements are not that precise.

    The McCrone report is not (yet) public. Note that it cannot (and does not) rule out a later fake. It just concludes (among others) that among all samples taken there is no trace of anything out of the time frame.

    Cheers, Rene

  53. Artur Sixto on February 22, 2011 at 7:59 pm said:

    Thank you Nick and Rene. 🙂

    If somebody feels like it, please have a go at the following cipher. I am not a cryptographer but tried to put myself in the place of a medieval paranoid alchemist. I took a natural language text and applied various simple rules to transmutate it in cipher. Wow!

    Will somebody find the philosophical stone to solve it? If nobody did better forget about the VMs!

    RO LEEQEI REE LRO NI EOLRCO EERNI CRCO EQO RIO NNRONI EO LRCREO CERPERLHO EEI I NI LI NO RIEBO LEEQO LO RORLO RENNO NNQIEENBO I NNI LRI EENYRPREO RLO ROLHO ROE NNONNQIEO NI EO LRELRPRO PI RBIO LEGBOQI NNE LRQEIENCO EINO EENO CE EO NRNRERGRII NCE LO RELEC CIOO RNI GI CIO LQIE NO RO BREQII RIENO LO NO NIROE NELEI O RCELRCORO RE LRNO NI ROLI RNERYOO LEGRECO LEGRENCRO NRENRI BLERO LRONO EENO NRCRENRO NI EO LRERPO CENCRIO RELRO NI EO LRCO EERNI CRPERO LCO NRI EPERO RENRE GERLO RPO NI RCO NCI CIO NRCRIE LRPERLO GO NYCE PRO CII REERELRO NI EO LRCO EERNI CRQO NO PO REI PERO GO QO RI NNLO CRIO CE LREPEEPLO BRO LQO NGERLO CO LRII RERO CO PNI RO NRQIE NO LERO QENEI O RELCO NNO CNEI LO CO NQI QENCIOO EBELRHI EO NRI NCE LRELEEE NNRQIEO GO QO REI PENELCO NNO CNEI LO CO NQI QENCIOE RLO CO PO CI NO NCE CO EINI COCIO NO EBE HO RO NNENI REENYRO GREBI QI NO NEERNO LERO NCIOO LO QI COE NRO CIENO NO LO HERO RQIIOE NCPEROO QIIE NRI NNEREBOE RPECIO LEENNLO CO POCI NO NCE COEINI CO CIOE NQIO LREQO LCO RO BRERQEEQIE LO RELEC CIOO RNI GICIOLO PERO CO PERLHO EE HO NI NGI NPE REGECNE LONO EENO CO HI EO NI NYOCIO CE LERE RPECIERCO EERNI CO CERHI EO NINYO CIO QIEO BI BO O LEPNREEO EBELRO NI EO LRCE CO EPO NYIO QIEE RCE QENENOI NENNI CREEE BRERCE LO GO EI LIO HO RO NPERNORO LNRERI RO EPE RERLO GECNEI LO CO EINI CO CIOE NNREO NI EO LRCE CO EPO NYIOI HI EO NROBI BOO LEPNREECE PEREENRE CO NO RLORNO EPROPI NRO CNEI CO NRI CEROCIO CE PEBO NER

  54. Artur Sixto on February 22, 2011 at 8:14 pm said:

    I forgot to say I’m not promising it’s 100% free of errors. One shouldn’t expect a mad alchemist to be infallible. But applying the magic formula in the reverse order should work.

  55. Artur Sixto on February 22, 2011 at 8:43 pm said:

    Sorry, please forget the previous challenge. I just tried to undo things and it doesn’t work because of a silly mistake. Still it might be interesting to see whether the “encrypted” text has VMs features like following Zipfs law, etc. If the VMs author screwed up things like me (or didn’t mind an irretrievable text to sell it as mysterious stuff) then it might be worth checking whether silly mistakes still produce undecipherable “cipher” with nice special features (Zipf’s law etc.)

  56. Artur: not as easy as it looks, eh? 🙂 If only it had been the magical cipher, thbt wpwld hbvf bffn fbsy! 🙂

  57. This rises a question. Is rubish “cipher” so easy to produce (on purpose or not)? It seems like I managed to thrash quite successfully my original information, but are there enough traces of a natural language and funny rules to qualify as a damn mysterious thing? It would be terrible if the VMs had structure without contents or without retrievable contents.

  58. Vytautas on February 23, 2011 at 7:28 am said:

    I am afraid about cipher of 3 layers (IMHO) – nightmares for some impatient decoders are guarnteed 🙂

  59. The VMs fits the profile of a fraudster’s book. I can picture some glossolalic magician “reading” from it. He could have invented a meaningless written language to complement his glossolalic skills. He could have believed enough in his special powers to craft the complex script and drawings. The work of an obsessive self-delusional magician or gone mad alchemist would look exactly like the VMs. It seems a mental case.

  60. Artur: it’s certainly not widely known that most real-world ciphers contain enciphering errors of one sort or another, particularly when they’re printed. Which throws up the issue that most ciphers are, well, partially rubbish, and that a decipherer has to see past the mistakes. I suspect part of the VMs’ tricksiness is that a lot of the structure is intended to give the decipherer cues & clues, and that probably 3%-5% of letters are miscopied. Actually, there’s a terrifically strong rationality threaded through the VMs, a set of guiding principles for the text, that give it shape and a kind of relentless cryptographic tempo: if (as an exercise) you transcribe a reasonable-sized page into EVA, you’ll see whole networks of patterns emerge.

  61. Nick, a telephone directory also has guiding principles, patterns and tempo. We just wouldn’t dance to its music if it was played. But while it isn’t and we are left guessing from a distance, one can dream of Ravel’s Bolero and dance to it endlessly. (Who would chose a directory to dream with?)

  62. So maybe they are chapter and verse references?

    Or references to hours, minutes and degrees

    or …

    By the way fol.67 contains a chart for calculating lunar position and phase. Just in case anyone’s curious about it.

    http://voynichimagery.blogspot.com/2012/01/fols-67r-ii-inner-motifs-moon-and.html

  63. xplor on June 1, 2013 at 3:36 pm said:

    Does the lack of religious sensibilities indicate a humanist author or a humanist translator?
    Compared to Ulisse Aldrovandi (Bologna, 1522-1605) a few generations latter it lacks Curiosities, Serpents and Dragons. Is it the work of Tutivillus ? Can the date of the last rebinding be determined.

  64. judyofthewoods on July 7, 2013 at 9:26 pm said:

    Could the “page numbers” be pages of another book, maybe a medieval one? Maybe those pages contained some kind of table or list which had to be refered to.

  65. Diane on July 25, 2013 at 3:22 am said:

    Another footnote to your post, Nick.

    If these strings refer to numbers of some other kind – such as quanitites, there’s plenty available about medieval European weights and measures, but I found this site which provides those used for different goods in medieval Cairo and, I presume, Alexandria, through which so much trade was done into medieval Europe.

    http://www.medievalislamiceconomy.uwo.ca/measures-egypt.html

    It’s a good site. Other pages there treat the broader issues of trade in, and with, the Islamic empire of that time.

    ‘D[irham]‘ or some such abbreviation might then reasonably occur before or after short-ish numeric series.

    With regard to herbal- and/or alchemical medicine, that dirham is the usual weight mentioned. For other dry goods such as cloth or ‘spices’ the ratl or rotl was a standard unit for the trade into Europe, as within Islam.

    “Ratl” is a term which some think derived from cylindrical containers of a kind which had earlier been used earlier hold scrolls – and which were originally coated with red. Very like the simpler containers in the Voynich ‘pharma’ section.

  66. Diane on July 25, 2013 at 3:25 am said:

    Xplor
    There’s no overall lack of religious sensibilities in the Voynich imagery, only absence from it of monotheistic imagery of the Christian or Muslim tradition.

  67. aiir
    and
    aiiv

    If the usual ‘v’ were meant for a ‘d’ – which it resembles in the MS, then it may refer to a dirham-weight.

    I can’t help it if the character is written in a way that sort-of, sometimes, looks like a Latin ‘v’.

    and if the ‘r’ were indeed an ‘r’
    it might refer to the ratl/rotl measure.

    Both are attested in documents of the 12th-15thC so no problems there. (They are earlier, and later too, which is cute).

    but the ‘a’ .. is it?
    and if it is, what part of speech?

  68. Ramon de Parma on November 22, 2014 at 11:08 am said:

    Hello. You ask which part of this argument do you disagree with?

    Sorry to say I disagree with all of it! Sorry!

    You start by saying these things look like medieval page references.

    You then say that ‘Whether you like it or not, there it is: the Voynich is utterly stuffed full of these medieval page references.’

    You then say they are NOT page references! Illogical!

    You admit that the elements that you find are also integrated into other words in the text.

    You then say that therefore the
    whole thing must be a cipher.

    I’m sorry to say that this whole discussion is not logical at all.

    But more important, it misses the obvious alternative explanation.

    All of this could be explained if the text is a human language written in an unknown script, not a cipher. Those elements you mention are just groups of letters, no more. They are nothing to do with page references.

  69. … so far as we know …

    Perhaps this should be an automatic caveat on every Voynich-related weblog comment.

    What do others thinK?

  70. Personally, I don’t think Nick’s argument is very safe. It is by no means the most parsimonious way of understanding the nature of the text.

    While the Voynich text could well be ciphered, this argument does not prove it is.

  71. Ramon: unfortunately, you’re missing the point of the argument completely. The Voynich Manuscript’s text contains a very specific family of signs – the ai[i][i]v and ai[i][i]r groups – that performs the function of page references within numerous medieval manuscripts.

    However, the way that these are used is not at all like actual page references – for instance, even though the Voynich Manuscript has more than ten quires, these apparent page references only refer to the ‘a’ quire (i.e. the very first quire): you will look in vain for any “biiv” / “biir” / “ciiv” / “ciir” groups in its Voynichese text.

    Moreover, as I argued in “The Curse of the Voynich”, “a–v” groups occur disproportionately often compared to the “a–r” groups: which is to say, these page references appear to refer to even pages far more than to odd pages.

    My conclusion then (as now) is that reading these as genuine page references is not tenable even slightly: and so the idea of reading the Voynichese text as merely an unusual or unknown language that happens to include medieval page references (but that refer only to the first quire, and then mainly to pages 2 and 4) is not tenable.

    As a corollary, if you want to argue that Voynichese is an unknown language, I think you have to conclude that these are not genuine medieval page references (even though they resemble them so closely). You can’t have both at the same time, that’s the whole point.

  72. Thing: you’re missing the specific point of this argument, which is that it acts as a disproof of the notion – as proposed by Ramon and numerous others over the years – that Voynichese can be an unusual plaintext that genuinely contains medieval page references in these groups. Hence it’s not a way of understanding the nature of the text so much as a way of demonstrating that the family of groups of letters that so resemble medieval page references cannot actually be medieval page references.

    As such, you cannot take these groups of letters at face value: from which I personally conclude that Voynichese is a cipher, because a ciphertext is a text that you cannot take at face value. 🙂

  73. “a way of demonstrating that the family of groups of letters that so resemble medieval page references cannot actually be medieval page references.”

    You’ve set up an hypothesis about page references and refuted it. Nothing more, nothing less. If you personally wish to believe that this proves the Voynich manuscript is a ciphertext, that’s fine, but you have not demonstrated it.

  74. Thing: no, what I’ve done is connected a recurrent pattern inside the Voynich Manuscript’s “language” with a large body of existing literature, and shown how that causes problems for people who want to take Voynichese at face value.

    The problem with Voynich researchers is that they so rarely find ways of connecting the things in the Voynich Manuscript with mainstream historical research that they don’t recognize it when it happens. So here it is.

  75. The connection is your hypothesis, and you’ve found reason why that hypothesis is wrong. You cannot argue that because the hypothesis is wrong the connection must thus have another meaning.

    It’s akin to saying, “this is water, but it doesn’t taste like water, so why does the water taste odd?” The correct answer is not, “it is something else made to look like water”, but rather, “I was wrong in thinking it was water”. Refuting your own hypothesis tells us nothing about your manuscript, only about your hypothesis.

  76. Thing: for those who study the Voynich Manuscript as an object entirely unconnected with Western codicology and palaeography, I’m sure your point is valid. But I do not: I study the Voynich Manuscript as an object completely embedded within Western codicology and palaeography, and so for me the systematic resemblance between the Voynich Manuscript’s aiir and aiiv groups and the aiir and aiiv page references of Western medieval manuscripts is something I cannot bracket away.

    Perhaps the real problem here is actually one of time, because the time it would take to build up a powerfully compelling literature-based argument based on this one tiny detail is beyond the amount of time I have available. But all the same, I’ve laid open the entire argument for anyone to examine and disprove: in eight years nobody has. Just dismissing the whole thing as hypothetical isn’t a particularly productive way of carrying on discourse: instead, why not look up medieval page/folio references for yourself and see what you find?

  77. Nick, I’m not arguing with your evidence but your logic. Particularly between 5 and 6. That is a terribly big leap to take and where I think your argument fails. I put it to you that concluding these characters groups must be cipher shapes is not the only “rational historical inference”.

    One alternative is that these character groups are a normal part of whatever process (possibly linguistic!) created the text. Here’s my evidence for this:

    1) As Rene mentioned on 21 February 2011, other members of this character group set don’t look like page references. The characters “ar” often occur as a word ending but have no place in the supposed page numbering scheme, nor do a whole slew of possible combinations.

    2) Many of the same character groups also appear in lesser number with “o” rather than “a”. Indeed, “oiin” is more common than “aiir”! But they too seem to have no place in your theory.

    3) Sequences of “e”, “ee”, “eee” occur showing that some characters can be double or triple, making there no need to describe “i” groups in a special way. Indeed, “e” occurs before “o” or “y” in single, double, and triple, and “i” occurs after “o” or “y” in single, double, and triple. And they don’t often occur together. Extricating a small portion of these patterns from the whole means that we miss the bigger picture.

    I know that your position is for the Voynich Manuscript to be a ciphertext, but you need a better argument if you’re going to convince others. If this is the best argument there is, then the question of cipher or language is wide open.

  78. SirHubert on November 22, 2014 at 11:18 pm said:

    Nick:

    I’ll have one last try.

    I know you posted this blog entry a few years ago, but are you able to give instances of a mediaeval text citing another book in this way (quire/folio/side)? Classical and scriptural texts tend to be divided into book/line for classical poetry (eg Aeneid ii, 328), or book/chapter or book/chapter/verse for prose (Thucydides ii, 65; Qur’an ix, 33; John 11, 35). Insofar as ancient or mediaeval writers did specify the precise source of a quotation rather than just assigning it to an author and sometimes also a work, I can’t imagine them using quire and folio numbers for this. It wouldn’t have worked – books were individually copied and produced, each with its own pagination and layout.

    I’m happy to be corrected (or simply told that I’m wrong, which is what usually happens here nowadays), but wouldn’t folio/quire/recto-verso be for bookbinders or copyists, rather than for readers?

    Oh, and for what my opinion is worth, I’m afraid Thing is dead right. As currently phrased, your chain of argument doesn’t prove what you claim it does. That’s a matter of logic and the structure of the argument itself, not content or context. I try very hard not to contradict you on your own blog, but perhaps you might have another look at this one?

  79. I’m hesitant to break into this animated conversation, but as someone who has had to reconcile the non-Mediterranean elements pervading the pictorial text with the Mediterranean materials and appearance, Nick’s discussion very helpful, and is well matched by other details and other documentary evidence – all of which together point to a conclusion that our fifteenth century object came from exemplars of earlier date, and non-Latin contexts, but was inscribed by persons who hands had been trained in the European Latin tradition. As you’ll know if you’ve had to copy a text you cannot read, it is difficult not to revert to one’s habitual way of writing any item which *does* seem familiar – just as it takes quite a while to write in a way that a native doesn’t immediately recognise as foreign. Crossed ‘7’s and so on. What Nick has done is point out that these groups look very like the Latins’ way of writing page references, but that they cannot be page references of that sort. In the same way, on a different folio, an expert agreed with me that the writing did look immediately as if it were meant for an inscription in Hebrew, but that the orthography was clearly not that of someone well accustomed to writing that script. It’s not such a surprise, since a third researcher – whose name Nick will doubtless recall, as I regret I can’t at the second – some years ago said that the script looks like someone ‘drawing’ writing. All these, and other things, are valuable contributions, each adding another brush-stroke to a true picture of the object. What is emerging pretty clearly is the picture of a non-Latin work carefully and patiently reproduced as best they could by people who couldn’t read a word of it. Whether this is because the whole is enciphered, or because the scribes’ reflexive use of Latin modes [e.g. ‘page references’] made it unrecognisable, or because there is still a hidden explanation (e.g. map-grids have two axes, and the day is easily divided into eight two-hour phases, and medicines may have been described in increments to four.. or tolls, or taxes.. ).

    In short, Nicks comment makes sure that no-one simply settles for a superficial reading of those ‘page references’, because he’s made clear that they are highly unlikely to be anything of the sort, so not much help as a hook for decipherment.

  80. PS SirHubert, your comment had not appeared when I wrote the above, so please don’t think it meant as contradiction of your own.

  81. Thing & SirHubert: there’s a missing step in the chain, sure – but that step is the business of doing (capital-H) history.

    When faced with a question such as this, methodologically the right thing to do is to immerse yourself in as much of the contextual material as you can, with the specific aim of building up a picture of what was going on in context – not the context of a 21st century person looking at an image of the Voynich Manuscript glowing on his or her PC screen, but the context of a 15th century person looking at the Voynich Manuscript itself.

    Though I formed an argument that makes specific claims, that came at the end of a very long process of immersion into the culture of medieval writing, delving deep into such works as the fascinating and densely informative books by the late Malcolm Parkes and many other palaeographers. For them, these medieval page references were an integral part of the landscape, something learned as part of reading and writing: understanding medieval manuscript culture without them was simply not possible.

    So yes, I might have another look at this one: but it would help me a lot if people bothered to read even a little bit about medieval writing culture before sniping at my arguments.

  82. SirHubert on November 23, 2014 at 11:18 am said:

    Nick, nobody is sniping at you. You asked in your original post: “Now, please tell me – which part of this argument do you disagree with?” and you have had a polite and courteous reply.

    Also, without asking you to go to a lot of trouble on a Sunday morning, I would be very interested to see an example of a mediaeval text where a textual reference cited by quire, folio and recto/verso as opposed to chapter/verse or similar. This isn’t something I personally have ever come across, and I’m struggling to find anything about this online, but I know you have spent a lot of time reading up on this field -which is why I asked you the question.

  83. I wish instead of sniping at other people’s work, each Voynichero would sit down and write a point-form list of their basic assumptions, then turn the chess-board around (as it were) and establish just how much objective support the manuscript-as-object provides for them.

    Such as:
    I assume the manuscript had an author.
    I assume the manuscript had a European author.
    I assume the European author came from …x… culture.
    I assume all pictures of plants must be like European plant-books and therefore about herbal stuff and European plants.

    Where in the world did you get that idea… no, don’t tell me. I bet you just followed the crowd at chow-time.

    That sort of thing has a medieval description: it’s called scholasticism. Logic increasingly divorced from fact and reason.

    (sour face emoticon)

  84. SirHubert: though many medieval (and indeed early modern) documents use medieval page references, one of the places I’ve seen it often is in late medieval herbals, where recipes and plants are often cross-referenced to each other. Unfortunately, a fair few of those were at the British Library (which has a super-restrictive copyright regime in place), so I’ll have to go looking elsewhere to bring up some examples for you.

    Many other medieval and early modern documents (remembering that folio numbering was only typical of the 16th century and later, though a few medieval documents do employ it) include an index of the contents, which is another place where you’ll commonly find medieval page references.

  85. SirHubert on November 23, 2014 at 12:10 pm said:

    Nick: Thank you – that’s exactly what I was hoping for. I’m sure there are enough reproductions of mediaeval herbals online that I’ll be able to find examples without you having to spend lots of time looking for me.

  86. oh dear –

  87. SirHubert on November 24, 2014 at 11:10 pm said:

    Looking back through the previous comments on this post , I’m clearly not the only person who was unaware that this kind of indexing was practiced in the 15-16th centuries, and that includes several wiser heads than mine. But following Nick’s comment I’ve put two-and-a-half hours commuting time to good use and found this:

    Keiser, G.R., “Scientific, Medical and Utilitarian Prose” (2004):

    “”The verse translation of De Re Rustica of Palladius deserves notice here for its particularly eloquent testimony to the concern for the practical value of such treatises. The translator…provided an alphabetized table at the beginning of the work; this contained arabic folio numbers and stanza letters corresponding to those found in the text. Given the ease with which scribes might follow an exemplar of a verse work, it seems likely that the translator intended the table to serve for all future copies.”

    “Surprisingly and interestingly, copies of alchemical writings consistently lack apparatus that would help a reader discover their contents easily. [BL, MS Harley 2407] does contain numerous illustrations, but these are of little values in locating specific treatises…Most manuscripts that preserve Middle English alchemical prose were compiled in the sixteenth and seventeenth centuries, and even these, copied at a time when the design of books had been more fully developed, contain little helpful apparatus.”

    Keiser is writing about British vernacular scientific works here, which may or may not be a directly valid comparison, and his folio numbers are in Arabic rather than Roman numerals. But it’s still the same principle.

  88. SirHubert: this (Middle English verse translation of the?) De Re Rustica sounds like an outlier, in that the quire+folio+side page reference style was the norm, rather than any Arabic folio numbering style. If you read books such as Flegg’s “Numbers Through The Ages”, you’ll see that Arabic numbers themselves didn’t really start to gain wide popularity until the second half of the fifteenth century.

    In general, the need for textual apparatus for cross-referencing became more apparent in the fifteenth century, and this is mentioned in Prager and Scaglia’s book “Mariano Taccola and his book De Ingeneis”: Taccola introduced a novel pictorial cross-referencing scheme, which I mentioned in Curse. But this goes above and beyond page referencing, which was already widely used in a variety of documents.

  89. Ramon de Parma on November 28, 2014 at 12:22 am said:

    Good morning Nick

    Coming back to this page I am surprised that the discussion got so hot, and that it upset you. Let me start by saying in reply to this:

    ‘it would help me a lot if people bothered to read even a little bit about medieval writing culture before sniping at my arguments.’.

    This sounds really unfair on all of us. I myself studied medieval texts for many years, in the Vatican library and in other places.

    I’m sorry to say that these medieval page references you mention are never in my experience of medeval texts included as part of other words or within sentences. So these parts of the Voynich text can’t be page references. I sorry to be so direct.

    So I agree with Mr. Thing and Sir Hubert as they say:

    ‘Refuting your own hypothesis tells us nothing about your manuscript, only about your hypothesis.’

    ‘As currently phrased, your chain of argument doesn’t prove what you claim it does. That’s a matter of logic and the structure of the argument itself, not content or context. ‘

    This makes me think that because you are the best known Voynich person who thinks the text is a cipher. And if your argument is not so strong then maybe it is not cipher at all???!!! I agree with this:

    ‘If this is the best argument there is, then the question of cipher or language is wide open.’

    So sorry that I am not sniping, only agreeing with these others.

  90. Ramon: the medieval page references I mention are indeed (in my own experience of late medieval texts, specifically herbals) included within sentences, but never within words. I’m therefore guessing that the majority of your own experience is of early-to-middle medieval texts?

    All of which is precisely why I find it so curious to find these distinctive, universally recognised shapes apparently embedded within words here. My inference from that is that they cannot actually be performing the same function as page references (which argues against this being a plaintext), and so that it is a cipher. But that is based on my experience of encountering medieval page references within late medieval texts.

  91. Why someone wrote a text, like this ?
    Why someone can not solve it, in such an modern era?
    if this text about herbals (may be partially), why encrypted?
    sophisticated and confused

  92. Ramon de Parma on November 28, 2014 at 4:23 pm said:

    No, my experience is in early and late texts.

    For me your obsession with these page references (which are there but not there!) is very strange.

    The best way to explain these shapes is as letter sequences, not ‘page references’.

    But I think we can’t persuade you. Do you know anyone who thinks the same as you? If not, maybe you should think it all through again?!

  93. Ramon: if you’re now saying these page references “are there but not there!”, we’re not actually in vast disagreement. 🙂

    EVA “aii” in Voynichese is overwhelmingly followed by EVA n (which is written as a ‘v’), EVA r (which is written as an ‘r’) and EVA m (which is written as an ‘r’ with a terminal loop)… and only very rarely by anything else. I’d agree that this might just be curious coincidence, and that Voynichese is a natural language written in a thoroughly artificial way… but I can’t honestly see this as anything but a telescopically remote possibility.

  94. John White on February 21, 2015 at 5:31 pm said:

    What about visual cipher… I’ve found that herbal on page f3v looks like “j” letter, herbal on f13v page looks like “Y” letter, herbal on f18v – like “L” letter, f18r – like “i” letter, f56r – like “p” letter, construction on f77r – like “F” letter, f78v – like “b” letter, f79v – like “L” letter, f81r – like “C” letter, f81v – “b” letter, etc. By the way, letter-shaped herbals mostly appears on “v” pages, so why “aiv”, “aiiv”, “aiiiv” groups appears much more than “air”, “aiir”, “aiiir” groups.

  95. John, do you mean to read the ‘r’ as recto, and the ‘v’ as verso?

  96. John White on February 22, 2015 at 6:53 am said:

    Diane, yes, i mean that

  97. Should read through all the comments before adding another “I can’t believe someone hasn’t already said this” but here it is – from the wiki article about the Apothecaries’ system.. of notation.

    “The use of Latin ensured that the recipes could be read by an international audience. There was a technical reason why 3 ʒ was written ʒiij, and 1⁄2 ʒ as ʒß or ʒss: The letters “ss” are an abbreviation for the Latin “semis” meaning “half”, while the Sharp S (“ß”) is an abbreviation for “ss”. In Apothecaries’ Latin, numbers were generally written, in Roman numerals, immediately following the symbol. Since only the units of the apothecaries’ system were used in this way, this made it clear that the civil weight system was not meant”.

    In other words, when an apothecary wanted to write “three drachms” (dry weight) the way to write it was ʒiij.

    wiki article “Apothecaries’ system”

  98. Thomas on June 22, 2015 at 5:03 pm said:

    Nick: The page references in a modern book are numbers, such as 7, 91, 15, 44, etc. My old school maths book has exercise problems in it, full with such numbers. For example 12 + 9 = ?

    I am sure these numbers in the text, i.e. the problems to be solved, cannot be modern page references. Therefore I conlude, (Q.E.D.!), that my school maths exercise book is written in cipher!

    No wonder I never solved it and still know little math! 😀

  99. Thomas: to be fair, you don’t seem to know much about medieval manuscripts either. 😉

  100. Thomas on June 23, 2015 at 6:23 am said:

    Nick: That is very true. I have never got near to or handled one in a library. I don’t need to know much about them when their script to be interpreted or translated is all over the place.

    I reassure you I am a sober dilettante. Once a pellingist, then turned baxist, then even hardcore cohenite, as the wind blows, while having a go at the immense task of cracking the mystery in my simple way. 🙂

  101. D.N. O'Donovan on November 18, 2015 at 1:03 pm said:

    Thomas,
    sheer curiosity – because I like your way with eponymous adjectives, what would you call someone who followed my views? An ODonovanite? A Dianovan? What?

  102. Diane: can you please try to avoid posting such trollbait on my site? Your comment is practically begging people to abuse you. 😐

  103. D.N. O'Donovan on November 19, 2015 at 12:29 am said:

    Not at all; the compliment and the curiosity are both genuine.

    I also doubt that Thomas would respond with such immaturity, and surely it is more important to curb inclinations to abuse that to encourage them by attacking the person asking a question – albeit an amused one.

  104. D.N. O'Donovan on November 19, 2015 at 12:45 am said:

    On a more serious note, Nick, would you mind if I offered my sympathy here, where it is more likely to be seen, to families and friends of three German backpackers who are believed to have lost their lives here in Australia yesterday?

  105. Rick A. Roberts on November 20, 2015 at 12:52 pm said:

    I started looking at the cipher that Artur Sixto posted previously. I started with the last line, since it is the shortest line. ” CEROCIO … NER “, = ” WAX THERE IS (ONE CELLED MANY SEEDED BERRY SUCH AS ; CUCUMBER, MELLON, PUMPKIN, or SQUAHSH ) KIDNEY “. Another line that I looked at is, ” LRCREO … RORLO “, ” BILE OCHER LERP (SWEET WAXY SECRETION FOUND IN AUSTRALIA AND TASMANIA ON EUCALYPTUS LEAVES. THIS IS PRODUCED BY JUMPING PLANT LICE OF SPONDYLIASPIS. IT IS EATEN BY ABORIGINES), TO IODINE NICKEL LITHIUM NOBELIUM AND THE DOWNY OAK “. The aborigines ” BUSH and TUCKER “, diet consists of six main nutrients; water, fat, protein, carbohydrates, vitamins, and minerals. The bile, found in most vertebrates, is used to digest lipids and break down fatty foods in duodenum of the small intestines. Bush Herbal Medicine uses the Eucalyptus Leaves to fill dental cavities. Most foods that we eat contain the elements of Iodine, Nickel, Lithium, and Nobelium.

  106. boyfriend , Champollion,,. :-) on November 20, 2015 at 4:06 pm said:

    No search at the manuscript alchemy. Not even search for the Philosopher’s Stone.

    Philosopher’s Stone : Atalanta fugiens . Michael Maier. Says :
    Make a circle, for men and women. Makes them square. Will make it a triangle. Make circle…… And you have the philosopher’s stone.

    It is a very easy procedure.
    The whole proces is a mathematical puzle. 🙂

  107. Out*of*the*Blue on November 20, 2015 at 7:33 pm said:

    There is an optical illusion on VMs f71r. It’s really just a simple example of artistic misdirection. Imagine the two blue-striped patterns on the page before any of the rest of the illustration was completed. Now see how their appearance and any potential, heraldic interpretations have been altered. These obvious, radial interpretations are superficial. However, the original, hidden, now secondary interpretations present a strong historical connection – supported by multiple internal elements in the illustration – to those who are knowledgeable of that history. And being in a secondary position is a clear indication that the disguise is intentional. Can the VMs author pull the wool over your eyes, again?

    The identification of a questionable pattern with a particular armorial insignia is always going to be a matter of speculation, particularly if there are multiple options among the insignia. But not all speculative investigation is necessarily rampant and unbridled. Speculation can be controlled by the imposition of certain criteria and parameters, where the facts are supportive. Two simple criteria limit the speculative possibilities seen here to a single historical event. Three objective positional placements on the same page confirm the specific identification. And after a fourth objective, positional confirmation in the Zodiac illustrations proves conclusive, the quibbles about speculation are irrelevant. Just a simple optical illusion that has taken a hundred years to see.

  108. Rick A. Roberts on November 21, 2015 at 3:22 am said:

    Looking some more at Artue Sixto’s posting, I looked at the line of cipher, ” NCI … EO “. I came up with, ” THERE WHAT AND KING NICKEL CHROMIUM BUT FOR PERLIS ( STATE OF NORTHWEST PENINSULA OF MALAYSIA ) CENSOR FOR CITIZEN ( APERT LOW TOWNSMAN WHO IS A PRAGMATIC TRADER ) BUSINESS AGES UNLESS MORE “.

  109. Rick A. Roberts on November 21, 2015 at 3:29 am said:

    Upon further looking at the cipher previously mentioned, I studied the line, ” RENNO … ROE “. I came up with, “KIDNEY THE STATUS NOT GOOD IODINE YEAR ? ? ENT? THEIR GUILTY “.

  110. Peter on June 22, 2017 at 5:39 pm said:

    @ all
    I am currently in a discussion about the Taurus in the VM.
    Is the last character now a “g” or an “o”?

  111. Peter: it is definitely ‘g’ (EVA ‘y’), it was misread as an ‘o’ many years ago when the only scans we had were very low quality.

  112. Peter on June 22, 2017 at 6:31 pm said:

    Thanks Nick
    For me it looks like a fold in the parchment.

  113. Peter: oddly, the three words by the Pleiades-like cluster appear to have been written with a slightly greener ink than the ink on the radial strips. All of which means that there may be something slightly unexpected going on here, so we should be perhaps remain wary about drawing easy conclusions either way.

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Post navigation