When I was young, I often used to play Scrabble with my grandmother Win on my way home from school. (By which I mean her maisonette was on my route home, not that we played Scrabble on the bus.) Which probably helps account for the deep-rooted enjoyment I still get from weird and wonderful words, many decades later.
From way back then, my favourite English word has always been “svelte” (though “tergiversate” was nipping at its heels for a couple of weeks last year). The reason I particularly like svelte is that it’s (I’m struggling to describe) ‘productively onomatopoeic’, in that the slow ‘l’-sound in the middle makes it feels elegant (indeed svelte) on the tongue. Really, it’s a word with an unusual (but nicely matching) mouth feel, one that manages to stand out from a dictionary sized pack. With getting too synaesthetic on you, to me it’s a kind of David Gower four of a word, a left-handed ping that’s over the boundary before the fielders even notice it’s gone. Something can’t be half-svelte, it’s either got it or it hasn’t.
Svelte also brings right to the fore the mad ragtag heterogeneity of English, the arbitrary coupling together of chance encounters over the millennia. To some it sounds Svedish Swedish (or perhaps a piece of stray Elvish?) but it’s actually a French word (svelte), from an Italian (svelto, “stretched out”), from Vulgar Latin (ex + vellere, i.e. to stretch + out).
(You might therefore suspect that it shares some kind of origin with “vellum” which is also stretched out, but the latter has its roots in “veal”, i.e. young calves: hence vellum is properly fine calfskin.)
Languages are like that: for all their modern apologists, academies, and syntactic niceties, they’re at heart accidental rather than designed. Esperanto and all the other modern conlangs are all very well, but a good part of the charm of real-world languages is the way stray and mongrel words hop in to fill the semantic gaps that inevitably open up as culture mutates and evolves. English obviously needed a word that expressed presence of svelteness in an object, why else would svelte have succeeded and persisted otherwise?
But (and isn’t there always a but in Cipher Mysteries)… where’s all that in the Voynich Manuscript’s language? Even if William Friedman was completely and utterly wrong about the Voynich’s being an artificial constructed language (which he was), I really can see exactly why he thought & believed that. For Voynichese words show such a strong family resemblance – a strongly interlinked productive grammar, if you will – that it almost precludes anything else. Whatever Voynichese is, there is definitely an artificiality to it, or at least an abundance of artifice. I suspect that anyone trying to map Voynichese onto a direct language base will almost inevitably find (to their eventual embarrassment) that it’s just too artificial to be workable: and that’s pretty much what Elizebeth Friedman concluded too.
So here’s your Voynich paradox for the day. I’m sure that there can be no “svelte” in the Voynichese ‘language’ as we see it, because the overwhelming majority of its words arise from a compact productive grammar quite unlike that of a real, heterogeneous, messy, accidental, historic language: and yet the look of Voynichese so resembles a language that it’s hard not to feel as though you’re perpetually a mini-dictionary away from just reading it.
Of course, for me the resolution of this paradox comes down to a well-chosen bunch of steganographic tricks (such as verbose cipher, shorthand, etc) that serve to conceal the plaintext in a misleading form… but you will no doubt have your own theories about how to slice through such a Gordian knot. 🙂
So, can you imagine just how svelte-ly a diplomat would have had to correspond with either their employer or friends?
😉
I’m also of the opinion that the VMS does not resemble a natural or otherwise synthetic language.
I would conclude that the VMS is either non-sensical or the result of some lossy encoding process – though i favor the latter.
In one test, i assigned a node for every word in the VMS vocabulary and let two nodes be connected iff a single character change (insert, delete or set) transforms one into the other.
In the resulting graph, we can study how connected the VMS vocabulary is by inspecting the largest independent subgraph and comparing the results with other languages.
Such an analysis of all VMS words indicates that 85% of them are connected. This means that, if we were to take two VMS words at random, there’s a (85% * 85%) = 72% chance that we can transform one into the other by successively replacing one of the words with another VMS word that differs by at most one character.
So, the VMS has a highly connected vocabulary – not surprising considering the perceptible amounts of repetition and structure in each word – but how does it compare with Latin, Italian and Hebrew texts?
Here are the results side by side – where the percentage value indicates the size of the largest word subgraph relative to the total number of words in the vocabulary.
VMS: 85%
– VMS-A: 82.4%
– VMS-B: 85.5%
Italian (Dante): 42.9%
Latin (Pliny): 17.1%
Latin (Bible): 16.0%
Hebrew (Bible): 80.5%
At a glance, Hebrew seems to have a sufficiently connected vocabulary to match the VMS. However, this changes dramatically if we focus on words of a certain length. Here’s the same analysis for words consisting of six or more characters:
VMS: 76%
– VMS-A: 67.1%
– VMS-B: 75.5%
Italian (Dante): 1.5%
Latin (Pliny): 0.5%
Latin (Bible): 0.5%
Hebrew (Bible): 1.6%
In the VMS, 76% of its six-or-more character words are related – according to the graph analysis. None of the other languages even come close to this.
The fact that, despite these results, some properties of the VMS, such as “word density” (i.e. the number of distinct words versus the number of total words) are closely aligned with the same values for other languages, is what suggests to me that the VMS is not merely a fake.
Here are the word densities for comparison:
VMS: 21.3%
– VMS-A: 29.9
– VMS-B: 21.2%
Italian (Dante): 21.0%
Latin (Pliny): 27.3%
Latin (Bible): 22.6%
Hebrew (Bible): 28.3%
Finally, it’s also interesting to observe how sorting the characters in each word impacts connectivity. Here’s the same analysis for six-or-more character words, but with sorted words:
VMS: 86.8%
– VMS-A: 78.0%
– VMS-B: 84%
Italian (Dante): 47.0%
Latin (Pliny): 35.4%
Latin (Bible): 23.1%
Hebrew (Bible): 16.95
Even sorted italian only goes up to 47%, not to mention that the VMS also becomes even more connected when it’s sorted. In order to further increase the connectivity of the Italian text we might have to start removing characters or replacing words with keys from an index – resulting in a lossy encoding process.
Thoughts? Nice blog BTW.
Dear all, I see one problem with Job’s pretty impressive analysis and that is the following: (1) I think some of us (Nick, if I can speak for you) are pretty certain that many words are missing glyphs. This is indicated by scribal abbreviations using their home-made “Tironian notes” for truncations or deletions. (2) word lengths are arbitrary, just made up by splitting up words or compounding simpler words. Some test decrypts indicate for example that “tion” occurs a lot. It is probably a suffix split off from a preceding word. I think (and I believe Nick will agree) that we have to crack the scribal abbreviation problem first before any decryption of the VM can succeed..
Job: thanks for sharing your results, I think they tell a very interesting story. Though numerous researchers people have remarked on how Voynichese’s words appear to have been somehow generated, I don’t recall seeing any similar study trying to quantify the degree of that overlap and comparing it with natural languages.
As for why this is, well… you say “lossy encoding”, I say “abbreviating shorthand”, but perhaps we’re ultimately talking about the same thing. There’s a very interesting pairs of papers by Mark Perakh where he concludes that some kind of abbreviation is going on, have you seen that? It seems like something you’d probably find both interesting and revealing. 🙂
PS: nice comment BTW. 😉
Are regular conjugations and interrogatory “wh..” words like “whither”, “who”, and “wherefore” art or artifice? If “sl…” words are most often associated with offensive things we will decide whether (kindred word, there) another “sl…” should fill in a semantic gap. Can we distinguish between art and artifice? The designed framework of language must have been created after the most basic utterances were established. The more unmixed a rudimentary language, the more effect the designers have on it. The evolution seems to be to ignore more and more of the design so we don’t have to think or know as much as we talk. And maybe we prefer art over artifice to boot. Echo: Where’s all that in the Voynich Manuscript’s language? I will guess it is either an arty artifice or an artful artifice and more difficult to reverse-engineer than a cipher. A lossy encoding process must include design to constrain glyph adjacency on esthetic principles. I think abbreviation is improbable because of the burden it places on the transformation.
I am not embarrassed by dead ends. As for disproofs, experience just takes longer to disprove our disproofs, if they are disprovable, than it takes to disprove our proofs. The only certainty is uncertainty and I am not certain about that, either.
Re. Job’s post. This is the kind and quality of work I like. I’ve plotted edit distances (using someone else’s script) without, however, getting back to precise quantifications. As he says, there are perceptible amounts of repetition and structure in each word. I think of that in terms of the density of the most frequent nGrams. I have seen nothing comparable. I am not sure the sorted-by-character words tell us much since longer words are more apt to be disqualified. I hope to see more of this.
deer Knox, I lik yur filosuphy! Well sed!
beedee
Job
It occurs to me that If you ran the same comparative tests using only one of the biblical books at a time you might get rather different results.
Not only because some books are genealogical, others narrative history, or poetry but because the collection was formed over a period of about 1700 years which is evident in the Hebrew original, even if not affecting the uniformity of Latin or Greek translations.
For interest, I went through the Brit.Lib’s online catalogues of manuscripts within the range 1340-1450, to see whether there were any manuscripts with the same dimensions as ms Beinecke 408. I was looking for traces of any standardised production of parchment and/or quires.
It’s more survey than strict sampling. I used all data from the illuminated mss catalogue, but only mss with a herbal section from the restricted-use catalogue, which has a more limited range of search options.
Even so, the results gave enough examples and these consistent with previous parameters established by the C-14 dating and work done by others on palaeography and codicology. Most of that matter having been published here by Nick, or through links and comments, I hope his readers might find this of interest too.
A couple of outliers are included, to show that the ‘160 mm’ seems to have been a standard as much as a century earlier.
According to the Beinecke’s catalogue, the Voynich measures 225 x 160 mm.
(voynich.nu. gives 23.5 x 16.2 cm (235 mm x 162 mm), but I adopt the holding library’s description).
key – Dimensions: page (textbox);Place of origin; Date; Language; Hand.
EXACT
Brit.Lib. MS Harley 2993 – 225 x 160 (145 x 110); Italy, N. E. (Venice) 1437 AD; Latin; Gothic.
Brit.Lib. MS Harley 5233 – 225 x 160 (140 x 95); England, c. 1436; French and Latin; Gothic cursive.
and
Brit.Lib. MS Harley 632 – 295 x 220 (220/225 x 150/160); England 1438 AD; Latin; Gothic cursive (Secretary).
Then, to see whether either measurement might represent a norm in a given time, I went through the catalogues again, taking each of the measurements separately.
______
PAGE has a measure of 225 mm.
(i) external
MS Yates Thompson 14 – 335 x 225 (215 x 130); England, E. (Mulbarton, Norfolk?) c. 1330-1340 AD; Latin; Gothic.
MS Additional 10456 – 300 x 225 (210/230 x 145/155); Eastern Mediterranean or Germany, 1349; Hebrew Ashkenazi;square script, punctuated.
.
MS Harley 5421 Pt2- 225 x 145 (150 x 90); Giovanni Boccaccio Bucolica; Italy, Central (Florence) 1408 AD; Latin.
MS Royal 19 B XVI – 305 x 225 (210 x 150); France, N., 1428 AD; French; Gothic cursive.
MS Additional 15423 – 330 x 225 (200 x 125); Italy, Central (Florence) 1441-1467; Hebrew; Italian semi-cursive script, punctuated.
On the offchance that the ms was copied from an exemplar but without wasting parchment on the original margins, I also checked measurements for the text-boxes.
(ii) textbox – 225mm.
Harley 632 – 295 x 220 (220/225 x 150/160); England 1438 AD; Latin; Gothic cursive (Secretary).
Harley 3949 – 285 x 200 (225 x 140); Italy, N. between 1447 and 1455 AD; Latin; Gothic.
Additional 16577 – 325 x 220 (225 x 140); Italy 3rd quarter of the 15th century; Hebrew; Italian semi-cursive script, partially punctuated.
PAGE has a measure of 160 mm.
(i) external
MS Royal MS 12 C XIX – 220 x 160 mm (text space: 145 x 90 mm); England, central or northern, 1200-1210; Latin and Old French; Gothic with tironian crossed et . ampersands. Anthology contains bestiary, Isidore, theological excerpts.
MS Additional 11639 – 160 x 120 (90/80 x 65/55; with marginal text: 100/90 x 65); France, N. 1277-1286; Hebrew.
MS Additional 15282 – 230 x 160 (145 x 100); Germany 1st quarter of the 14th century; Hebrew, Aramaic; Ashkenazi square and semi-cursive script, main text is punctuated, marginal text is unpunctuated.
MS Harley 6613 – 160 x 105 (110 x 70); 1st quarter of the 15th century, after 1409; English; Gothic. Lollard text.
MS Additional 18970 – 160 x 110 (90 x 60); Spain 15th century; Hebrew; Sephardi semi cursive and square script. punctuated.
MS Harley 2698 – 160 x 115 (105 x 80); Italy, S., 1423; Latin; Semi-humanistic.
MS Lansdowne 344 – 160 x 115 (110 x 80); England (Midlands?) c. 1425; Latin and English; Gothic.
MS Burney 289 – 235 x 160 (150 x 100); Italy, Central (Florence) 1427; Latin; Humanistic.
MS Harley 5233 – 225 x 160 (140 x 95); England c. 1436; French and Latin; Gothic cursive.
MS Harley 2993 – 225 x 160 (145 x 110); Italy, N. E. (Venice); 1437; Latin; Gothic.
(ii) text box
MS Additional 9405 – 250 x 190 (160/165 x 125); Germany 1309; Hebrew; Ashkenazi square script, punctuated & semi-cursive script,unpunctuated.
MS Additional 9406 – 250 x 190 (155/160 x 125); Germany 1309; Hebrew; Ashkenazi square script, punctuated.
MS Harley 5613 – 220 x 145 (160 x 90); Eastern Mediterranean, May 1407; Greek; Greek minuscule.
MS Additional 60577 – 210/215 x 135 mm (text space: 160 x 90 mm); c 1487-1574; French & Latin, Middle English. (includes herbal folios)
MS Harley 632 – 295 x 220 (220/225 x 150/160) [see above ]
-and that was all –
Diane: an interesting survey, thanks very much for sharing it!
Perhaps a study on the dimensions of Quattrocento herbals might show a trend, I don’t know if such data have been collected anywhere in a useful form.
Nick,
I’ve called it a survey because a proper study would be a Master’s thesis at least. Each library has its areas of specialty and preferences for which will be digitised and so on. Plus there’s no search by dimensions in the Brit.Lib’s more restricted catalogue online. But then again, I found it very interesting that not one work of the northern Latin tradition, and only one of the Byzantine came up.
As for herbals – yes, it’s true that a huge part of the ms shows plants. What I tend to think is that other aspects of the manuscript should make us wary of assuming that the content will conform to that limited genre simply because it was the most prevalent in Europe.. or at least the genre best represented in our remaining manuscripts.
The Vms is, after all, a compendium and there are many other applications for plants and their extracts.
I guess, though, if I were interested in the text I’d be curious to test it against the Macer Floridus. Popular especially about the time and place I think this matter reached the west.
Oh, btw. If I were into genealogy, and in Europe, I’d be finding out if there were any connection between one “Shaeffner” who was the first person to publish Walfrid Strabo’s devotional-gardening work ‘Animae Hortulus’ aka the Hortulus and that pharmacist-physician Shaeffner to whom Jakub Horckiky was apprenticed. But alas – circumstances prevent.
Two more. Physic/herbal works – Brit Lib’s Illuminated mss online.
Same typical ‘160 mm’; none with 225 mm.
I’m a little disappointed; had hopes of seeing the Lombardy Herbal (MS Sloane 4016) among them.
Brit.Lib. MS Harley 2558 England, S. W.; 1st half of the 15th century; Latin and English; Gothic cursive
210 x 150 (160 x 105)
and
MS Sloane 1977 France, N. (Amiens); 1st quarter of the 14th century French; Gothic. 235 x 160 (160 x 110) in two columns
~ yadayay ~
Knox,
Certainty is working for me. I’ve certainly translated (not decoded–there is no code) 15 Vms folios. I’ve also tentatively identified the writer of the entire manuscript, if not just the botanical section. I’ve translated several of the “balnealogical” folios as well. One particular balnealogical page/folio does not identify those objects which “look like” maracas. What it does, is describe the effects of those “globes” — which immediately identifies them as being mandrake fruit.
I do admire your observations and point of view: different sections of the Vms require different points of view with which to consider or ponder — and reach a conclusion.
🙂
Diane, i agree, i would expect this type of graph based analysis to also vary according to the contents, rather than just the language.
On the other hand, i imagine it would be difficult to find an unencoded, or naturally occurring text that can match the VMS on this metric – where 76% of all words, with six-or-more characters, are related.
Learning Voynichese sounds like a nightmare, with so many words that look so similar. How does Chinese compare?
Thomas, are those results well established? From my analysis it seems at least plausible that the author might have used a procedure that operates on each word separately, thereby maintaining the original word boundaries – for example, the values for “word density” in my previous post seem aligned with other languages – but this could be coincidental.
Knox, i explored the results for sorted-words because i believe it would be one of the more effective ways to increase the word connections – besides removing/replacing characters.
Nick, i will try to find those papers, i’m curious because to be honest i’m not sure abbreviations can account for some of the properties of the VMS. For example, there seem to be around 4500 words in the VMS vocabulary that have five or more words that differ by a single character.
BTW, by “lossy encoding” i meant any procedure that introduces ambiguity – e.g. such that even the author might have to stop and think about what the original value was – IMO a shorthand may fall in that category depending on how it’s executed.
Finally, since my previous post i’ve also noticed that, for the most part, but not strictly speaking, the most frequently occurring words are also the ones that have the most one-character variants.
For example, nearly all of the words that have no one-character variants also occur only once. The more often a word is used, the more variants it is likely to have. This may not be unexpected in a language, but i find it suspicious, given the large amounts of single-character variants in the VMS, that this association should exist.
Job – it does sound very condensed-recipe like, doesn’t it.
e.g. medicine
Hot and Dry in the [1-4] degree = HD[x]~
or dyeing
Mordant [Alum,Tin etc] gives Yellow = M[x]Y
or weaving
(1st) Yellow,Green, Red, White, Black =Y.G.R.W.Bk
(2nd) Yellow,Green,Blue, White,Black – Y.G.B.W.Bk
even musical, and navigational .
language of gradations?
Job: remember that in the 15th century (to which most of the internal construction evidence points), there were two types of shorthand: (1) a tiny handful of special symbols (word-final “-9” meaning -us/-um, word-initial “9-” meaning “com-“/”con-“, etc) hanging over from Tironian notae, and (2) local scribal abbreviation and contraction systems. Modern shorthand systems begin 150+ years later.
In Voynichese, I’m pretty sure that “-8-” means “contractio[n]” and “-9” means “truncatio[n]”, both of which are from the latter kind of shorthand (but, in the case of the -9, masquerading as the former). I suspect that these drive a lot (but far from all) of the stats you’re looking at.
In August of last year, I offered an identification of the plant in vms f-15v: Squash/cucurbit and a full translation (into latin) of the accompanying discussion (based on Ububchasym’s method). It was that folio which enabled me to begin fully translating the botanical folios,
Even today, if an “edible” plant is being portrayed, I look for those five “qualities”. I also go to the “pharma pages/folios” to find the recipes using combinations of various roots and leafy materials which have been previously portrayed.
So, I was able to translate the scribal “language/shorthand” ciphers into intelligible latin. Elsewhere on the WWW, one can find other references to scribal notations — of which I’ve found only a few which apply to the Vms/Boenicke 408.
The most commonly used combination of scribal “shorthand” is “aes”-“geus” — 89. This combination ends nearly every botanical discussion.
The elaborate “P” which begins any botanical item discussion, is automatically read as “Especies”
If there’s an expert on possible nuances and double-talk in seventeenth-century Latin, I’d be glad to know if in ‘diplomat-ese’ a phase of Baresch’s to Kircher might be meant to have a secondary reading in addition to the overt one.
Baresch wrote
… herbae peregrinae, in Volumine depictae, notitiam hominum in partibus Germaniae subterfugientes.
Neal’s translation
… the volume contains pictures of exotic plants which have escaped observation here in Germany.
What I’d like to know is it there might be second intent – to convey the idea that the book had been written by people who had to leave Germany without being noticed.
Any kind soul willing to enlighten me on the point?
Diane, a name you can check out is Paracelsus. I am trying to find the origin of his knowledge.
Xplor
If you google ‘Paracelsus’ with ‘Voynich’, you should find a fair bit.
oh look –
“There are several ideas concerning the etymology of the ancient toponym Adria/Atria. One theory is that it derives from the Illyrian (Venetic language) word adur, ‘water’, ‘sea.”
I never knew Venetic was Illyrian.
I am currently checking to see if the manuscript was owned by Tycho Brahe and sold to Rudolf.
An interesting book is Follies of science at the court of Rudolf II 1576-1612 . It is old and online.
xplor: I’m sorry to have to note that Brahe is a suggestion that has been extensively mined in the past by Voynich researchers, yet has produced nothing of value.
Xplor,
You’re not wandering too far off course, though, when you mention Tycho Brahe. Some of his writings were, at one time, part of Rudolph II’s correspondence and collections of oddities (such as the eyepiece for a telescope) and the “magic lantern”) — which eventually ended up in Father Kircher’s archives received as a consequence of the “Thirty-years War” and, maybe, the “Hundred Years War”.
You mentioned “Follies of Science…..” : I have been trying to remember the source of that publication for months now. I THINK I can now refer to University of California, but was not able to cite which campus!
I’ll be 70 y.o. in September. So forgive me if I lead you up blind alleys anywhere amongst my offerings to various discussions!
ps: I vaguely recall that two other practitioners (Follies of Science) were a gentleman named Dodoens, and the other gentleman being Clusius. Both were alluded to be court physicians to the Bohemian and Austrian domains.
Nick, please correct my “half-baked” memory of several hundred sources of information I’ve delved into over the last year or so!
I don’t think I am yet a candidate for Alzheimer’s, but it does take me longer to recall the names of some of my favorite people and or acquaintances.
Beady-eyed wonder
Nick, Diane, Xplor:
If, while perusing the various Voynich volios, you don’t find the “usual” 89 combined syllable/sound aes-geus or aes-keus, but DO find what looks like a tiny numeral 9 which “head/loop” sits on the line but extends to the back of numeral, you are seeing a combination of the ‘8’ and ‘9’ sounds into the sound ex-eus. The tiny 9 sound can be used for words such as ex-ercise, ex-am-ine, ex-ellent, —–
I’ve x-plaind these “Vms” syllables on several other discussions here and there. While I’m typing these words, my “spell-checker” is going “batty”!
🙂
Nick,
& then ve haf vords like sveetheart, Sveden, Sviss, Sven……
🙂
It is a well worn path as it should be. He was a seeker of truth, had a state of the art alchemical labs, had massive herbal gardens with over 300 trees and a large collection of rare books.
The problem is where did he get the book. He knew everybody and anybody in academic circles.
I can suggest a stenographic trick to cut through the Gordian knot!
Let’s consider the top right-hand image of 86V – the one which has an image of a castle with swallowtail battlements.
This looks to me like it could require a cylindrical mirror to be placed on the centre so that a less distorted image of the scene is reflected in the cylinder. This is called anamorphosis, and its just possible that it’s important to helping to decipher the Voynich manuscript.
Because not only the pictorial elements of the overall scene get transformed by anamorphosis, but the writing also.
As yet I won’t be putting it on my C.V. – but I think its got promise.
A few more notes on the anamorphosis possibility…
Apart from the most common shape used for anamorphosis – i.e. a cylindrical mirror – there are some other possibilities, such as a cone (either reaching a point or not, and either larger at the base and smaller at the top or vice versa) or a sphere (or near-sphere such as e.g. a goldfish bowl) or a hemisphere (either way up).
It’s also possible that the purposeful distortion (assuming there is a purposeful distortion) be corrected not by a mirror, but through something transparent – for example instead of a cylindrical mirror using a glass cylinder with water in it, or a small ‘goldfish bowl’ with water in it.
A somewhat less likely possibility (but a possibility nevertheless, in principle) is that of an anamorphosis device that allows the inside of a larger cylinder to be seen.
In this version, either a complete or a partial topless and bottomless cylinder would be used (think of e.g. a round biscuit tin with the bottom removed and the inside polished), perhaps just bigger than the text written ‘in the round’ between the carefully drawn outer and inner circles (which would project that text at the base of the reflected image, or at the ‘top’ if you had to look at the result ‘upside down’ for it to be intelligible), or perhaps of a size that falls between the outer and inner circles, since such carefully drawn circles might be guidelines as to the required size of the anamorphosis device for it to work.
Note that a long thin rectangle of steel, highly polished, could be flexible enough and reflective enough to bend to the curvature required and reflect the scene well enough to see and read. Although it would only reflect a portion of the scene at a time, it would be sufficient to move it round to complete the viewing. In fact a device of this nature would have the advantage – compared to a complete cylinder – that it can be adjusted for different sized images.
Fortunately it would be possible to study these possibilities today without turning up at the Beinecke Rare Book & Manuscript Library with a large assortment of cylindrical, conical, hemispherical and spherical objects of different sizes and in both mirror and glass versions (and water to use with the glass versions) – because the type of image transformation necessary has been perfected relatively recently in computer software, especially software used for making and manipulating photographic panoramas. So between the high quality images of the Voynich Manuscript available online and the appropriate ‘panorama manipulation’ software, this avenue of possibilities could be explored relatively easily.
Oh, ChrisB!
I’m hoping Nick is looking at this right now. Because I have been referring to that six-folded page as the “Nine-Rosettes” folio, for lack of being unable to get a “folio number” from Boenicke. Also, I have been unable to download a large enough image to read the “captioning”.
Nick uses that “swallowtail-merlin” castle image at the top of his blog page. Fr. Kircher identified that structure as “Vellitrae” Now, all I have to do is find a cylindrical mirror! In the meantime, take a look at the “rosette” immediately to the left. The upper-case letters may spell PRAENESTE (a temple in Frascati).
I will be searching for a cylindrical object with a mirror surface: salt shaker, lipstick tube, pill bottle……..?
Strictly speaking its not 86V – its 85v1, 85v2, 86r3, 86r4, 86r5 and 86r6…..
A possible identification of that castle as written in that folio would be the phonetic spelling of oe-ll-tl-e:
oe = V (as in V-ellitre)
l-l = el (as in V-ell-itre)
tl = tl or a “tapped R” or “trilled” TR as in V-ell-tr-ee
I did find a polished stainless steel cylinder which I will be able to use, maybe, if I can get strongly magnified images of the folios from “somewhere in cyberspace”!
bdid1br – sorry, but I couldn’t see PRAENESTE in the top centre “rosette”. To be honest I can’t even distinguish between upper and lower case ‘letters’!
bdid1br,
here’s a place in cyberspace where you can get your very own strongly magnifiable image of the “rosettes” page:
Google “Beinecke” and choose the top result listed.
Click “Search” and then do a search for “Voynich” and again choose the top result listed.
Scroll down a bit and click on “View all Images”
Click on the blue text “Voynich Manuscript Cipher Manuscript”
On the next page, the upper image is clickable-onable to bring it up in its own window, but at first it shows the front cover, which is of little interest, and we need to change it for the “rosettes” page.
Below, you’ll see that he lower image has a slider that lets you progress through the document…
…so slide through the pages until you see what they call 86V, and click on it, which brings it to the upper window.
Now click on the upper window and 86V launches in its own zoomable and panable window.
But it also has an option to “Dowload image”, which will download a high resolution .jpg called 1006231.jpg.
So, Nick:
Is this latest discussion maybe developing into a rather “svelte” contribution? If I were to attempt to write “svelte” into Voynichese, it might look something like this:
?oelt :
Cyrillic sibilant which looks like a question mark, no dot
oe or ue are common Latin and or Greek symbols for “V”
The Vms character which looks like two poles, each linked with a loop, equals ell or el (as in v-el-vet)
The Vms character which looks like two poles linked, but with only a loop on the right-most pole equals the sound tl, as in atlas — but this character can also be used as a trilled or tapped R. What is very interesting to me is that I have yet to find a Vms character which would represent the sound of “l-t” as in built or build.
One other mysterious symbol, which I have found only in a few Vms discussions, is a scribal abbreviation which may be “tius” or, maybe “teus”. An example of this last character can be found on folio 55v, at the end of line 5. (my translation, to Latin, of that last word: oll-ax-a-ius (maybe itius). (Nevertheless, Vms f55v is the Water Lotus (Sacred Bean of Egypt).
Believe me, KrisB, when I say “been there–done that”! Many times, I’m able to zoom way-in on any section of a page. BUT when it comes to getting a print of just that particular zoomed section, my efforts are “doomed”.
Are you beginning to get an idea of my “tongue-in-cheek” humor, here? Ennyway…I have had many hours of enjoyable puzzle solving with Boenicke Ms 408 (for which I thank you very much, Nick) and WW-Webbers wherever!
Although I tossed out Mary D-Imp’s ms a long time ago, I still follow any discussion in re Currier’s or Tiltman’s efforts. Now that I’ve found the connection between Busbecq’s correspondence with his boss Ferdinand I (Austria), and Busbecq’s correspondence with Rudolph II’s (Bohemia) physician, Clusius, I may now be able to make even more coherent translations of Boenicke Mss 408. (The connection I’m referring to is current-day on-line discussions of Busbecq sending tulip bulbs to Clusius. In Leiden University archive is a collection of HANDWRITTEN letters to-from Busbecq and Clusius. I’ve only had time to read one, and because of the language used, and the photostatic (very outdated) process of conserving the record (I am NOT complaining) — it takes me a while. In the meantime folks, do yourselves a favor — and compare Ogier Ghislain de Busbecq’s handwritten documents with the document we call Boenicke 408.
I’ll be returning to the “pharma” section to see if it is Busbecq (diplomat) or Clusius (doctor) who may have written the “pharma” recipes.
for now, ciao! 🙂
For completeness, here are rendered versions of some of the graphs i described previously.
In these graphs, nodes represent distinct words in the text, color coded by occurrence – white nodes correspond to words that occur only once while dark green nodes correspond to words that occur more than fifty times.
Arrowed edges indicate that two words differ by a character insertion/deletion while diamond edges indicate that two words differ at a single character position.
For the VMS i’ve included versions with no edges because some of the graphs are quite dense. The images are between 3Mb and 6Mb each.
All words:
VMS – http://voynichms.appspot.com/images/vm-word-graph.png
VMS (no edges) – http://voynichms.appspot.com/images/vm-word-graph-noedges.png
DANTE – http://voynichms.appspot.com/images/dante-word-graph.png
Words with six or more characters:
VMS – http://voynichms.appspot.com/images/vm-word-graph-sixers.png
VMS (no edges) – http://voynichms.appspot.com/images/vm-word-graph-sixers-noedges.png
DANTE – http://voynichms.appspot.com/images/dante-word-graph-sixers.png
Additionally, here is an interactive version of the VMS all-words graph which displays word values when mousing over nodes:
http://voynichms.appspot.com/images/vm-word-graph-labeled-noedges.svg
These images were generated using Neato (http://www.graphviz.org/category/graphviz-terms/neato). The layout engine’s clustering algorithm seems to split the VMS into two fairly-dense, somewhat-connected word groups – i’m not sure this is particularly meaningful but it’s a noteworthy curiosity.
Forgive my earlier mispelling –ChrisB. Yesterday and early this a.m., I went back online with LeidenU’s fabulous archive (now gif files) of the documents generated by Busbecq and Clusius. But now we can get past the earlier “photostat”-black memoranda and supposedly download/print any and all of letters being presented.
I am now convinced that the commentary which appears on the Vms folio 116v (reference to Ancyrara and also reference to an ancient monument to Augustus) was written by Busbecg while on his diplomatic “tour” of Suleiman the Great’s domain. I am also convinced that Busbecq sent the tulip bulbs to Clusius — because Leiden University’s archive says that was so.
I am going to try to give Bill Thayer the “heads-up” if he is still maintaining his website at the University of Chicago.
Spell-check correction:the ref on 116v is “Ancyra”.
Diane: Busbecq was deliberately attempting to make his travel diary appear to be as innocuous as possible, and make it appear to be a “naturalist”s sketchbook with commentary — and even a naturopathic recipe book.
I’ll be going back to Leiden’s archive to attempt to find Clusius’ “side of the story”. Diane, thanks for the reference to Nick’s discussion of “nihil obstat” on folio 116v.
Diane & Nick,
I’m now going to order a copy of a book published not too long ago (supposedly translated) about Busbecg’s Letters. (Amazon.com?)
Correction to my earlier post in re Ancyrara: correct spelling is Ancyra (which is reference to Ottoman-controlled Ankara).
Obviously I appear to be very excited with my repetitious references to Ancyra. Here are some more references to Busbecq, Clusius, Suleiman, Bayezid, Tamerlane……a book written by Mike Dash (who writes articles for The Guardian, The Daily Mail, and The Fortean Times):
“Tulipomania –The story of the Worlds’s Most Coveted Flower & The Extraordinary Passions it Aroused” (pub.2001)
My long-awaited copy of “The Turkish Letters of Ogier Ghiselin de Busbecq” finally arrived yesterday. One can almost put the Voynich manuscript side-by-side with “The Letters” and match the translated words of Busbecq’s travels, plant identifications, and work in Suleiman’s domain and court. For example:
“Letters”, pp 68-70 refer to a cure for plague: “scordium”. In my collection of botanical books, I was able to find the plant identified as “dioscorea”, common name for yam. So, now I will be searching the Voynich to match the plant, the roots, and leaves, and the mixtures which appear in the pharma and recipes folios. I’ll keep y’all posted.
So, I shall continue my translation, folio by folio….
Diane, I hit the “post” button too soon: “Letters” pp 50-51 discuss a garment called “watered camlet (mohair), made from the hair of goats”.
Busbecq goes on to describe what appears to me, anyway, the process of felting (and possibly dyeing simultaneously in the Emperor’s favorite color of green).
Oh, no more cross-commentary. Have I upset everyone’s apple cart? I truly meant no harm or contradiction to other’s posts. My first instinct is to facilitate transmission of information (by whatever means) rather than contradict or argumentate. Bandwagons aside, Diane (other posts, people) I feel that I was born to be an educator/facilitator of “reading & spelling”. Never made it that far, even though I was local spelling champion, nine years old, for the state spelling championship competition (grades 1 through 12). Ten years old, and my favorite author was James Michener. Babble-on Beady-eyed Wonder (b-d-i-d 1-dr) 😉
Busbecq’s travel companion was his physician, Quacquelben. It is seeming, to me, that the “botanical/herbal” discussions and illustrations, and recipes MAY have been Quacquelben’s notes. Whether he, himself, wrote them or rather dictated to a scribe (who may have been a member of Busbecq’s traveling party) still remains to be “seen”.
bd1dr – please clarify
..Diane (other posts, people) I feel that I was born to be an educator…
I don’t know if you’re instructing me to read other blogs (other posts, people)
or if you’re instructing Nick’s readers to hunt through these posts to find my comments.
I expect no such attention if the last, and if the first your comment might benefit from moderation.
I suppose there’s some third reading possible, so look forward to hearing what it is.
Job
re permutations and variations by one char
once more the sort of thing expressable by acronym + number
sorry the example’s so late.
bibliotheque.rouen.fr/repons/portal/armadillo/image/5823..
Diane, i suppose any form of abbreviation or numerical notation is likely to raise the number of variants.
There is more research to be done here. For example, the number of variants per folio fluctuates quite a bit.
Folio 75r has a vocabulary of 195 words, of which 75% are related.
Folio 77r has a vocabulary of 169 words, of which 21% are related.
Is this the result of fewer typos, nulls, numbers or abbreviations in Folio 77r? Are they encoded with slightly different methods? (they are both B) Do transcription errors accentuate this property of the text? By how much?
The global value for the entire Manuscript is 85%, but we still need to quantify how much each folio contributes to this value – for example by excluding each individual folio from the text and recomputing.
The following additional folios resemble 75r in the excessively connected vocabulary:
Folio 116r 71.3%
Folio 103v 73.7%
Folio 80r 71%
While some nearly adjacent folios look much more conventional:
Folio 114v 24.3%