Letters hidden in Voynich plants…
For years, numerous Voynich researchers have pored over the VMs’ confusing images, hunting for any tiny clues that might possibly be hidden beneath the clumsily-applied paint. And yes, I admit that I’ve done probably more than my fair share of this kind of thing (Curse pp.96-102 stands as testament to this endeavour): so it’s now interesting to hear that René Zandbergen believes that there is “rather strong” evidence (a) that some of the plant parts in the VMs have letters to direct the colouring of that page (let’s follow Vera Segre Rutz and call them “colour annotations“), and (b) that those colour annotations are written in German. Here’s what René says:-
If we go back to the [alchemical herbal] web page of Philip Neal, one of the herbals mentioned there is “Vicenza, Biblioteca Bertoliana MS G.23.2.3 (362) s. 15, Italy and Germany“.
Actually, “G.23.2.3” is the old shelf mark and now it is usually referred to as “Vicenza MS 362“. It’s a 15th century Italian herbal, with illustrations from the alchemical herbal tradition. It also says ‘Germany’ because Segre Rutz in her book quoted by Philip describes that it has ‘colour annotations’ in German. Indeed, in the few illustrations I have from this herbal, you can easily see many occurrences of ‘rot’, ‘gr(ue)n’, ‘gelb’ (red, green, blue) and also ‘erd’ (earth) or ‘weiss’. Additionally it has one illustration with alternating red and green leaves, with alternating single ‘r’ and ‘g’ characters written inside.
These all look extremely similar to the few colour markings in the early quires of the Voynich MS. There’s a clear ‘rot’ in the root of f9v (already seen by many), there’s the ‘g’ in f1v, and then there’s another ‘rot’ with some individual ‘r’s under the paint of the viola tricolor mentioned above, and another ‘g’ to the side of the flower on the right.
Here’s one page of Vicenza MS 362 with ‘rot’ in the root (barely visible in this resolution), while this page has alternating ‘r’ and ‘g’ characters (and the same problem).
Well… the first thing to note is that if Vicenza MS 362 is an Italian herbal with German colour hints, then the same could well be true of the Voynich Manuscript – if we’re talking about a manuscript with depictions of Italian castles (there are actually two sets of swallowtail merlons on the nine rosette page), a provenance that (currently) starts in Prague, and Stolfi’s putative “heavy painter” adding organic-looking paints later in the VMs’ life, then there’s no obvious conflict between the two narratives. All of which begs a whole constellation of questions, such as:-
- Can the letters hidden in the Voynich’s plants be proven to be colour annotations?
- Can the letters hidden in the Voynich’s plants be proven to be written in German?
- Can the letters hidden in the Voynich’s plants be proven to have been written by the author?
- Can the letters hidden in the Voynich’s plants be proven not to have been written by the author?
- Can the letters hidden in the Voynich’s plants be dated independently of the VMs itself?
- Can the letters hidden in the Voynich’s plants be proven to have been written by the author of one of the marginalia?
You already know what I’m going to say: let’s look at the primary evidence for ourselves. With the help of Jon Grove’s clever colour remover plugin (which I recently discovered can work inside IrfanView, a highly recommended hack!), you can make a pretty good stab at seeing past the paint to the letters beneath. Here is an image of what I found when I tried this on the top left flower on f9v (the viola tricolor page I mentioned recently):-
Note that even though I tried really quite hard to image process this to a satisfactory state, I didn’t really succeed to the degree that I’d hoped: but still, I think it’s good enough to see that (a) the claim that the top left petal clearly says “rot” doesn’t really hold up; (b) that there are colour annotations in at least three of the five petals in the same hand; and (c) that the colour annotations are even smaller than the Voynichese text.
Let’s move on to the claimed colour annotation in f4r’s root. In “The Curse”, I built a whole cryptogrammatic superstructure on top of my reading of this (rotated) as “TOA”, which is one possibility (though in retrospect, I do see how it seems a little hopeful). However, I also think that reading “rot” here (as a column of letters) is just as hopeful but in a completely different way.
However, if you assemble these two (claimed) colour annotations onto the same page and add in the f66r and the f116v marginalia, I think you find something really very surprising indeed:-
What I’m claiming is that they all seem to share the same unusual-looking “topless p with a preceding upstroke” first letter (even the supposed “portas” word, which I’ve never believed was the right reading at all). What does it mean? Was all this added by the author or by a later owner?
Update: here’s the leaf Voynichese painted over in f2r René mentions in his comment #1 below (but with the green paint removed), with the rather puzzling EVA “ios.an.on“:-
February 27th, 2010 at 3:32 pm
I’m not a paleographer, and I hesitate to argue about the definitive
reading of the characters in the viola tricolor, but there seems to be
little doubt possible about the reading of ‘rot’ in the root of f4r.
The key, really, is in the parallel with Vicenza MS 362, which also has
‘rot’ in roots of some plants (one I have seen), and which is from the same
time as the Voynich MS and also from N. Italy. It is one of the alchemical
herbals which Toresella has compared to the Voynich MS.
Of all the marginalia in the Voynich MS, these are the most likely to be
from the original author / scribe.
The really, really interesting corollary of all this is, in my opinion, the fact
that there is also one example of Voynichese writing in one leaf in
the same part of the Voynich MS. It is distinctly possible that this
says ‘green’ one way or another. I won’t claim that it is certain, but
it is certainly quite possible. Thus, turning (Eva) ‘ios.an.on’
(tentative reading) into green could be the key to solving the Voynich MS.
February 27th, 2010 at 11:50 pm
I think it highly unlikely that that stands for green – why use 7 characters on page 4 to represent what was represented by 1 character on page one – much more likely in my opinion that he broke of here temporarily and scribbled that to remind himself of something knowing it would be covered by paint later.
February 28th, 2010 at 9:03 am
By the way, better images of Vicenza MS 362 are available on the net, but
they are ‘hidden’ by the fact that the MS ID for them is incorrect. They
are sold as posters of MS 320 M. Here’s a link:
http://www.bridgemanartondemand.com/art/142289/Ms_320_M_Fol31_Herba_Poleximas_from_Liber_Herbarius_una_cum
The images here can be enlarged by clicking on them. Under
‘related images’ there are a few more.
Some googling will turn
up even larger images.
The library also doesn’t understand where this invented “shelf mark”
comes from.
February 28th, 2010 at 10:20 am
The sequence ‘ios.an.on’ (or something similar) is unusual ‘Voynichese’.
If the author started by making drawings before writing text, this may
also have been one of the very first words written.
But let’s explore a little (yes, it gets speculative now). What if Voynichese
is indeed like a verbose cipher, and also like a universal language at the
same time. The author could have imagined inventing a script that
could be used to write all languages. The characters describe the
shape of the letter. Thus (just as an example), ‘ios’ could
represent a handwritten ‘g’, ‘an’ a handwritten ‘r’ and ‘an’ a handwritten ‘n’.
Then it would encode the (abbreviated) ‘grn’.
This is close to what Elmar Vogt proposes at his blog.
This would go a long way towards explaining some of the features
of the Voynich MS:
- it will make it look like a verbose cipher
- it gives rise to a word structure with the an anomalous entropy
- it explains the ‘line as a functional unit’ as character shapes of
a source text at start and end of lines would be different
- it is easy to imagine how different source texts lead to A and B languages
Exploring this further would seen extremely interesting.
It couls also be a complete waste of time ;D
February 28th, 2010 at 2:31 pm
Leaf f2r: Are you sure it’s voynichese? It looks like “iolan[th?]on-” to me. Close to the greek iolanthos = violet (the flower). The plant doesn’t look much like a violet, but hey, maybe it was someone’s incorrect guess.
February 28th, 2010 at 8:26 pm
To Rene: this is a very interesting conjecture (if I’ve understood correctly) that the VMs words code for letter shapes. I did look into something similar, namely that they code for letters and groups of letters in various plaintext languages, here, but didn’t get very far. If the VMs codes map to single letters, and there are around 10,000 unique words in the manuscript, we would have too many letters – so that cannot be what you mean, exactly?
March 1st, 2010 at 11:21 am
If you include Nick’s “layered cipher” theory to the stroke theory, I believe it explains why there are more different Voynichese words than letters in ordinary alphabets.
March 1st, 2010 at 11:30 am
Julian, in this ‘though experiment’ I was thinking that on average 2-3 Eva
characters could represent one plaintext character. This is not at
all a new idea of course, but it would be one way how the relatively long
string inside the leaf could mean ‘green’.
March 1st, 2010 at 1:36 pm
That would of course depend on how you spell “green” (and in what language).
March 1st, 2010 at 1:42 pm
Oh, and just to muddy the waters yet further, I’d point out that it looks to me as though both of o’s in “ios.an.on” have been emended (as well as painted over). Also, I’d transcribe it as “ior.an.oin” too (i.e. “r” rather than “s”, and “oin” rather than “on”).
March 1st, 2010 at 2:18 pm
Yes, as I already indicated in comment #4 it is not straightforward to
read. At this magnification, image artefacts (possibly enhanced by
the post-processing) start playing a role.
Whatever langauge the ‘green’ is in, it doesn’t seem to be the main
language of the Voynich MS, it being so atypical.
Thus I further speculate that the Voynich MS main text is not in
German (or rather, not a rendition of a German source text)
March 1st, 2010 at 8:12 pm
There is a nice “B” in the leaves of the plant on f39v, which I show on my (new) blog.
Since it appears in an unpainted area, I propose it is an instruction that the area is to be left blank.
March 4th, 2010 at 10:01 am
Yay, this is my field! Etymologically it does make sense. Blank (or cognates thereof) exists or existed in most western Indo-European-derived languages, and at around 1400 it evolved the meaning of “empty space” in English. It was borrowed into English from Old French, where it also had a meaning of “white” or “colorless”, so a word reminiscent of “blank” did exist in most Western European languages in the Quattrocento and it also bore a meaning which would make sense when applied to this instance.
March 4th, 2010 at 11:59 pm
Good grief … and I thought I was joking
March 5th, 2010 at 12:10 am
[...] is an interesting discussion of letters in plants over on Nick Pelling’s blog. Leave a [...]
April 7th, 2010 at 7:27 am
I’ve sometimes thought the 3-twist root thingy-s look not unlike a mark that appears on Khazar pottery… They seemed to like using bows and arrows, too.
March 14th, 2011 at 11:33 am
[...] blog post by Nick Pelling (with a discussion including René [...]
January 3rd, 2012 at 5:32 am
There’s a pun built into the picture on fol.1v which only works in a handful of languages: those which derive their word for cloves from the Latin ‘clavus’.
I won’t add a link to my own site though, unless Nick thinks it’s worth adding.
February 5th, 2012 at 11:21 am
Perhaps it’s on a different post – but surely someone thought to suggest that the letters refer to the pigment which can be got from the plant/s concerned?
A particular green (e.g. for the Hajii’s turban, or for vestments etc.) could explain the one-off form.