Voynich Quire Numbers

Cryptographically, the Voynich Manuscript (AKA ‘the VMs’) is a 234-page handwritten document filled with enciphered text in an unknown alphabet, threaded through by a large number of drawings of equally uncertain meaning – unrecognizable plants, circular diagrams, small naked women. It has inspired a surfeit (if not a plethora) of web-pages, articles, books, documentaries and even novels. Yada yada yada.

Yet the VMs is arguably just as great an historical mystery. Because its plausible provenance goes back only as far as 1608 or so (there is an erased signature on the first folio, using a person’s title that was only granted then), the question of what happened before 1608 remains wide open…

1. Dating the Voynich Manuscript

Two pieces of evidence help us to date the VMs: firstly, the presence of parallel hatching in some of its drawings implies a likely earliest date of 1450 (the date when parallel hatching began to appear in Venetian drawings, or perhaps 1440 if in Florence), while the 15th century handwriting used for the quire numbers implies a likely latest date of 1500 (as first suggested by John Manly in 1931). According to Graham Flegg’s “Numbers Through The Ages”,

It is striking that the old forms of 4, 5, and 7 are retained into the fifteenth century, at least in Central Europe. Then around 1500 the new forms, which have hardly changed since, became established almost overnight. (p.122)

Furthermore, there is very persuasive codicological evidence that before the quire numbers were added, the VMs had already had the order of its pages scrambled and had been incorrectly rebound by someone who didn’t understand its contents: this would arguably move the likely original date to earlier than 1500, say to 1480 or earlier.

So, unless there is strong evidence to the contrary, the codicological and palaeographical starting point for study of the VMs should be that it dates to between 1450 and 1500, probably favouring the earlier end of the range: and so the popular claim that it is some kind of late 16th century hoax therefore seems inconsistent with the basic art history (by roughly a century or so).

One researcher has an even more specific view: when Italian medieval herbal expert Sergio Toresella examined the VMs, he had the following to say (posted by Jim Reeds in 1995):- 

The VMS is, with certainty, authentic; not a fake.  It was manufactured
in the period 1450-1460.  It was in France for a while: the month names
on the zodiac diagrams are in French in a French handwriting.  The book
itself comes from Italy; the mysterious writing is done in a round 
humanistic style found only in Italy in the second half of the 1400’s.
There are similarities between the organization of the VMS (including 
the balneological section!) and that of other Italian herbals of the 1400s.

[ http://www.geocities.com/ctesibos/voynich/precednt.html ]

Unfortunately, this is very much the point where mainstream historical research into the Voynich manuscript flounders, with many unknowns still in play. But perhaps some historical white knight will now ride to our collective rescue by taking a fresh look at its unusual quire numbers…

2. The Voynich Manuscript’s Unusual Quire Numbers

Crucially, the VMs’ quire numbers may yet give us an insight into its very early provenance. These are of a form that has yet to be matched in any other document, a curiously awkward hybrid of Roman and Arabic numbers: match these, and we suddenly stand a chance of determining one of the manuscript’s early owners (perhaps even its earliest owner).

Even though the original quire number writer was thinking in Latin ordinals (primus, secundus, etc), what s/he was actually writing was an ugly mixture of Arabic numerals and late medieval -9 Latin abbreviations: pm9, 29, 39, 49, 5t9, 6t9, 7m9, 8u9, 9n9, 10m9, 11m9, [12 missing], 139, 149, 159, [16 missing], 179, [18 missing], 19, 20. Palaeographically, we can split these up into four hands:

The four quire hands, from The Curse of the Voynich (2006) p.17

To my eyes, Quire Hand #2 looks (from the even ink flow across the ‘m’) to have been written with a fine metal nib by a later owner (aping the original numbering scheme and handwriting): while Quire Hand #4 also looks to have been added later (though by someone with no grasp of how the numbering scheme worked).

Were Quire Hands #1 and #3 added by the same person? Even though they are broadly similar 15th century hands, my best guess is that they were added by two different people – the “3” seems different, and Quire Hand #3 is generally scratchier and finer. All in all, my best guess is that Quire Hand #1 was the original quire numbering hand.

However, the big historical mystery is why no match to this unusual quire numbering style has been found in any other document. Can it really be that this cipher manuscript has had an entirely unique quire numbering style added to it, or might there actually be examples of similar numbering (perhaps not even used for quire numbering) in other documents, perhaps even by one or other of the same hands that appear here?

All suggestions welcomed (both for documents to examine and for sources to refer to)!

18 thoughts on “Voynich Quire Numbers

  1. Patrick David on October 21, 2013 at 1:29 am said:


    Just got my full reproduction of the Voynich manuscript at the link above, i thought i would share since one has not been available for a long time

  2. Limoges on January 25, 2015 at 11:41 pm said:

    I don’t think that’s a 9. I think it’s an a with a diaeresis or a gra sider or an accent. They occur extremely frequently throughout the mss. text, especially at the ends of words. Thus, the majority of what you depict is “ma,” which could be short for mata, from the Old Norse meta, proto-Germanic metana, meaning a measuring, or marke, from Old Norse merki, proto-Germanic marko, meaning boundary/marker. I rather doubt the rest are numbers, either, but rather selected because they look to us like numbers.

  3. Limoges,
    Not quite sure that I follow. You say that you don’t see the quire numbers as numbers – so far so good.

    Then suddenly you leap from saying there are numerous forms similar to the ‘9’ in the manuscript – as there are – to saying,
    “the majority of what you depict is ‘ma’ ”

    I really don’t think that statement is supportable. And if you’re relying on etymological dictionaries of English.. forget it.

  4. Limoges: quire numbers – ordering systems written in a consistent place on the outside of a set of gatherings to be bound as quires – were a constant part of the mainstream European manuscript tradition for centuries. While it is possible that the Voynich Manuscript’s “quire numbers” merely resemble actual quire numbers and yet were something else entirely, the quires ended up being largely bound according to those quire numbers: and whoever bound them 400+ years ago (when the foliation was added) agreed almost entirely with the reading I give here.

    In fact, I don’t think it’s entirely unreasonable to hypothesize that the quire numbers may well have been added close to Lake Constance: http://ciphermysteries.com/2012/04/15/the-lake-constance-voynich-hypothesis But that’s another story. 🙂

  5. Limoges on January 29, 2015 at 7:33 pm said:

    Hi, Nick. You furthered the thought beautifully. The quires may certainly have been bound much later and the so-called markings taken for numbers, since a constant part of mainstream European manuscript tradition for centuries was to order and bind as quires. But that’s the cart before the horse. A lot could have been done to this manuscript to fit it into mainstream European manuscript tradition. Seems in another post you yourself raise excellent doubts as to whether the quires are in original order or even if the ink is original. I would question if we have enough information about the script to see the markings as a curiously awkward hybrid of Roman and Arabic numerals, as the above post suggests, especially since the “9” cannot be distinguished from a letter frequently used within the text to make words. Your Lake Constance hypothesis is intriguing. About five hours north is the Horselloch of interesting notoriety and five hours east is Prague, around which five medieval castles in accurate order resemble those of the rosette map. So whether or not that 9 turns out to be an “a” or in fact a “9” or something else entirely, the research is promising.

  6. Gelida on January 30, 2015 at 1:26 am said:

    Limoges –
    please reassure me that you’re not entertaining any idea of a shooting script which has the Voynich text a forbidden account of Tannhauser’s revels with the ‘nymphs’ of the Horselloch!

    We’ve just begun to get the study of this manuscript moving from “I wish…” scenarios to something more like an analytical history. It would be a shame to return to the bad old days of plots as silly as any opera cycle’s.

  7. Bertin on June 24, 2016 at 11:17 am said:

    “an ugly mixture of Arabic numerals and late medieval”???
    “though by someone with no grasp of how the numbering scheme worked”
    Are you serious?
    Do you really need to insult the person who wrote those signs?
    Because you don’t know why this person wrote those signs that way doesn’t give you the right to talk about it like that…

  8. nickpelling on June 24, 2016 at 11:37 am said:

    Bertin: if you wrote a series of numbers – “i ii iii iv v” – and then someone else completed the series “6 7 8 9”, I think it would be fair to say that the other person had “no grasp of how the numbering system worked”. Yet that would seem to be exactly what we see in the Voynich manuscript’s quire numbers: a series of abbreviated longhand Latin ordinals “pm9, 29, 39” etc being completed “19, 20”.

  9. Diane on June 24, 2016 at 11:46 am said:

    since the binding is – according to Touwaide – probably Italian, isn’t it more reasonable to say that whoever wrote the quire numbers *might* have been educated in the vicinity of Lake Constance? I find the ‘lake constance’ argument is less convincing than internally-consistent. As soon as one asks if a similar style has been sought with the same assiduity in the corpus of Italian, French, English, Spanish, Mallorcan or Frankish-Aegean manuscripts there’s a resounding silence. The librarian you contacted, as I recall, told you that such quire-numbers were not as rare (‘specific’?) as we were led to expect.

    Also – I’m surprised to see that you now think the Arabic folio numbers are also fifteenth century. I had understood that it was accepted they were rather later, and that a specialist in John Dee’s handwriting had even said he considered them written by Dee – presumably in the mid-to-late sixteenth century. I haven’t seen anything by way of evidence or reasoned argument to the contrary – do please refer me if another study has produced different results.

  10. nickpelling on June 24, 2016 at 12:24 pm said:

    Diane: I understand your reasoning re Lake Constance, but I somehow doubt that ‘probably Italian’ binding is as geographically precise as your argument would require it to be. When we uncover more examples of abbreviated longhand Roman ordinals, I would hope that we can start to make more progress with this particular vein.

    The librarian I contacted suspected that we might well find more such numbering in the St Gallen manuscript collection… but given that this has only a small number of 15th century manuscripts, we’re only talking about a dwindlingly small number of mss at best.

    As you know, the Voynich’s folio numbers use the kind of mainstream mid-16th century Arabic numeral forms that were used by Dee, Kelly and hundreds of thousands of other people, so I don’t know what gave you the idea that I dated them to the fifteenth century.

  11. Diane on June 24, 2016 at 3:05 pm said:

    Oh yes, I see on re-reading your response that there’s a tacit idea that some considerable interval occurs between the time the quire numbers were written and the time the manuscript was bound (the time to which you ascribe the Arabic foliation).

    I expect the evidence which led you to that view is in your book. I’ll try again to order a copy to replace the lost one.

  12. Diane on June 24, 2016 at 3:11 pm said:

    The Compelling Press site is still rejecting my efforts. 🙁

  13. nickpelling on June 24, 2016 at 3:36 pm said:

    Diane: http://www.compellingpress.com/voynich seems to work just fine from here – does clicking on the ‘earth’ icon at the top take you to a Paypal screen?

  14. nickpelling on June 24, 2016 at 3:38 pm said:

    Diane: if the quire numbering style can be dated to between ~1450 and ~1500, and the folio numbering style to the decades around 1550, and they appear to be written in different inks with different quills by different hands, the chances they were written close together in time would surely seem to be fairly small.

  15. Bertin on June 24, 2016 at 7:11 pm said:

    Is that you Diane O’donovan?
    Well yes the 19th quire number and the 20th are pretty strange. But “ugly” is little too much for me…

  16. nickpelling on June 24, 2016 at 8:30 pm said:

    Bertin: ah, I’m pretty sure I used “ugly” to describe the “abbreviated longhand Roman ordinals”. Any numbering system where “29” means either “second” or “29” depending on the context could be beautiful only to its parents. And perhaps to the lab technician who did the in vitro bit. 😉

  17. Thomas F. Spande on June 24, 2016 at 10:18 pm said:

    Dear all, Voynichers should be aware that what we all call “arabic” numbers are not arabic at all but come out of India and “Indo-arabic” is more fitting. The concept of “0” was an Indian innovation. Sir Hubert should be consulted on what “arabic” numbers actually look like. Just a punctillio as it only changes the vocabulary, not the numbers. Cheers, Tom

  18. Diane on June 25, 2016 at 3:27 am said:

    My fault. In the first instance I misinterpreted your saying “the quires ended up being largely bound according to those quire numbers: and whoever bound them 400+ years ago (when the foliation was added)..

    It seemed that you were saying the quiration, binding and foliation were near contemporary with the inscription of the quires (i.e. 1405-1438), which was a little startling. But your idea is that the individual quires were filled then, but only later (near the end of the century) given quire-numbers, and then rather later still (late 16thC?) bound and given Arabic folio numbers.

    Must say it all seems a bit hypothetical .. have we any qualified person’s opinion that the binding (stitching) looks late 16thC?

    It all depends on when the quires were bound, doesn’t it. Suppose the stitching is characteristically 15thC and Italian?

    Other hypotheses could be spun. How about… between 1405-1438 a pre-bound blank manuscript (of the sort we know were produced, and especially useful for note-takers and student) was filled with matter from various source-works.

    Quire numbers were written in, since lack of an index or foliation made it that much easier to turn quickly to the wanted passage or page.

    Later again (16th century?) a third person thought this silly and added folio numbers, though not exactly getting them right.

    One might add an extra elaboration – a binding or re-binding from the earlier text(s) as reason for quire-numbers to be added. But as I say, it all depends on the sitching’s date, doesn’t it?

    About ‘paypal’ – I expect the problem is two-fold, the small and select (read ‘minor’) bank I use, and the fact that I do not provide my telephone number unless I wish to be telephoned. I’ll try abebooks.

Leave a Reply

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