I am having a really really hard time restraining myself...
I tried. I really tried. I saved a response to the translation as a draft THREE TIMES and said to myself, "Don't post this yet. Don't post this." I didn't want to discourage the blogger from posting his system or showing us more "translations" because I thought they would be interesting to see, but I kept looking at this translation and, well, how can one NOT comment on it?
And then I thought, but a detailed response might provoke the blogger to change his system, so then I thought, should I post it as a spoiler? but realized that wouldn't make much difference. Everyone would look at it.
So, I have decided to email it to one person and not post it, at least not yet. I am too eager to see the next translation installment to spoil the fun.
I must admit it is entertaining to read.
So, Fun with Words it is.. Mein Brevier, Bring it On!
That really long german word mentioned in the 'Keys to the Vms' article is missing a z at the end. But as a born Ösi I can't but counter with
Donaudampfschifffahrtselektrizitätenhauptbetriebswerkbauunterbeamtengesellschaft
Note that this word gained an additional awkward f in the last orthography reform. For practical use, someone looking for a Donaudampfschifffahrtsgesellschaftskapitänsanwärterposten would have employed abbreviations and code words, like "I geh DDSG hackln".
And because it starts with terrible dog I need to apply some „Es laut gar vbel, vnd man heisst es Küchen Latein, so man Latein redet nach aussweisen der Teutschen Zungen.“
Sorry, you kind of need to understand german at least a little bit to fully enjoy them:
Whenever I see a new solution proposed, I think:
Ovum, ovum, quid lacus ego!
Which is mostly followed by:
Nunc habemus Endiviam. Nescio quid mihi magis farcimentum esset.
And whenever I see a latin solution proposed I can't help but think pseudo:
SITA VSVI LATE IN ISTA PER CANES
and this makes me kind of..
Caprum non iam habeo.
And because this is all a bit about cloisters, Μὴν ἦπτε οἵ; ῏Ηπτε μὴν νι οἵ· ἦπτε μὴν γράς
which makes sense (sic! - for german readers, haha).
On a more serious note (I'm trying to stay OT, whatever this abbreviation may expand to), there seems to be a consistent expansion system which I need to look at later.
But actually I'm more in need of good laughs today, so I can't spare you with the BSR, Berliner Stadtreinigung (cleansing service) and their real fun pseudo way to call things, like "Tonnosaurus REX" for their largest dumpster truck (tonne = rubbish bin), and the dustbins (Mülltrennung is big in Germany) in the botanical garden:
Glasus vulgus
Papyrus antiquus
Plasticus berlinensis
Restus wegwerfus
Have a nice day.
Apparently EVA "an" encodes "Anus", 118 times or so.
Make of that what you will.
Most of the words so far aren't Latin, as claimed (nor is any of the grammar), even though the glyphs that follow Latin scribal conventions are being expanded with valid Latin prefixes and suffixes.
Google Translate might interpret them as such, because the software makes a "best guess" as to what mystery words might be, but taken individually, they're not Latin.
The "translation" really falls apart with some of the longer words (as might be expected). You can combine almost any short syllable with a valid Latin prefix or suffix and get something that might be a word (as can be done with Greek, Turkish, or Arabic), but one is less likely to hit a "bingo" with longer words—for those the method needs to work.
What do our Latin scholars think of "ilaque aris aferanius" (translation of line 1 f1r)? Even if you correct "ilaque" to "itaque" it doesn't mean anything, and aris doesn't fit the context (and aris probably only works because it's easy to create words by adding valid Latin suffixes to short syllables).
Searcher has a much better grasp of Latin and made a heroic effort to translate the VMS into Latin, a creditable effort, I would say, but ran across all the usual VMS stumbling blocks in terms of repetition, glyph-position inflexibility, etc. From what I've seen so far, Lockerby doesn't appear to have a working (or even rudimentary) knowledge of Latin and would thus find it difficult to judge (or even recognize) whether it were sense or nonsense.
I'm still withholding judgment until I see more, these are very short phrases and one needs to see a whole paragraph to really know, but so far, half the words are not Latin and there's no grammatical sense to it.
Now the full manuscript has been translated:
You are not allowed to view links.
Register or
Login to view.
It is the same old gunk of "Latin" as we've seen so many times before. I applaud the blogger's effort but I'm a bit disappointed. I really thought he was onto something initially.
Next!
It's somewhat refreshing and bold that he generated a 'translation' of the whole manuscript. This makes it really obvious where the problems lie with approaches like these. One clear problem is "eeee" series: "oeeeees eteoesanus esteeum conlieteeum apolietum etaquees"
I would be interested if someone had the capability of computing the character and digraph entropies of this 'plain text'.
My prediction (based on the fact that the Voynich text is being 'expanded') is that the digraph entropy of the plain text is even lower than that of the Voynichese text, which is already anomalously low.
Lockerby has translated the first word as "peraetumes" and then says it is to be understood as "'peractum', a book for teachers and students".
Peraetumes or even Peractumes is not a Latin word.
He does not explain how he gets from "peraetumes" to "peractum" (other than saying that some of the ees are supposed to be cees), and peractum does not mean a book for teachers and students, as he claims. Depending on context and grammatical form it has a variety of meanings such as disturbed, harassed, completed, and would not be followed by "con ilaque" (ilaque is not a word in Latin, the only word that's close that is a correct Latin word is "itaque" but then all the ells in the transcript would have to be amended to tees for it to be consistent).
If you glance through his transcript, what you see is a lot of repetitious short syllables with Latin (and French) endings added to make them look like Latin. It's like taking a lot of nonsense syllables in English and adding "-ing" and -ly" and calling it English:
"umlio estoque este iloesesteum feretqueum esaesaus esaus ileumeses"
"etoque estoesaus ilestum iletum esorus etoesanus esto iletois"
"umeteeum oilaum oilum esanus oileteum iloilanus"
"feretestanus estoileteeum etoque liestoeseeesum"
Even if you change some of the ees to cees, there's not a single Latin word in these four lines, except for "esto" or maybe "este" (which is more French than Latin). One correct word every few lines (usually the short words) is not a translation.
To call this Latin stretches credulity pretty far.
All this is a result of working 'backwards', i.e. to try to convert the Voynich text to plain text, instead of the other way round.
When one sets ones mind to thinking in the correct direction (i.e. converting from an unknown plain text to the Voynich text), one will quickly realise that the table of translation that is proposed makes no sense.
One has to turn the translation table around.
This means:
- occurances of 'cus' were encoded as Eva-b
- 'per' as Eva-f
- 'cis' as Eva-g
(skipping)
- 'ap' as Eva-q
etc. etc.
These are just a very minor subset of possible Latin bi- and trigraphs.
What a coincidence that these occurred so often in the plain text, that they deserved to be rendered by a single and frequently occurring character.
One of the most common mistakes in all proposed solutions.
Quote:I would be interested if someone had the capability of computing the character and digraph entropies of this 'plain text'.
My prediction (based on the fact that the Voynich text is being 'expanded') is that the digraph entropy of the plain text is even lower than that of the Voynichese text, which is already anomalously low.
I could do that with the scripts You are not allowed to view links.
Register or
Login to view. described, if anyone please provides me with raw
files (only Voynichese, spaces preserved, no linebreaks, no punctuation). No need to prepare the entire VMS, some 10k characters would do.
Btw, I'm confused whence Patrick produces the punctuation?