+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Analysis of the text (https://www.voynich.ninja/forum-41.html)
+--- Thread: Computational Attacks on Abbreviated Text (/thread-2046.html)
Pages:
1
2
Computational Attacks on Abbreviated Text - -JKP- - 16-08-2017
D. O'Donovan [corrected] posted an excerpt of Latin text on her blog and included the interlinear expansion of the abbreviations You are not allowed to view links. Register or Login to view..
![[Image: LatinExpanded.png]](http://voynichportal.com/wp-content/uploads/2017/08/LatinExpanded.png)
Praepositio* et praeterea per dyptongon scribitum. Pene vero, quod est coniunctio, per E. Pena quod est supplicium per OE Q. littera tunc recte ponitur, cum illia statim .U. littera sequitur et alia quelibet una plures ve vocales iunguntur. ita ut una syllaba fiat. Cetera per .C. scribuntur. Que [Quae] pronomen cum .A. scribendum. que coniunctio sine .A.
[*I'm not sure why the scribe has written this as Pre perpositio as I'm fairly sure Prepositio/Praepositio was intended.]
My version is not intended as a criticism of the one posted on O'Donovan's site, I just feel it should be as accurate as possible if it's going to be discussed on the forum (and I enjoy expanding the abbreviations).
I've been wanting to write in more detail about the way Latin was expanded in the 15th century, and have touched on it in some of my blogs, because it does dramatically change the statistical properties of word-frequency and other computational attacks, but can't seem to find enough time, so I thought this short excerpt might be enough to provide a start for a thread on the computational properties of medieval texts (note that the example above is about three centuries before the VMS).
[Latin abbreviations are old news to some members, but if you are not familiar with them yet, Cappelli is an excellent resource.]
Even if you study Latin abbreviations, and attempt to break the VMS text out into Latin (as has been tried by a number of researchers, and as has been recently attempted by P. Lockerby), that doesn't mean the correspondence between VMS glyphs or glyph-combinations is consistent. One can see in this short excerpt that "quod" was abbreviated in two different ways. It was not uncommon for words to be abbreviated four or five different ways. By the 15th century, long after the above excerpt was written, handwriting was messier, abbreviations more frequent, and consistency in the abbreviations even less than one sees in the above example.
These abbreviation systems were not limited to Latin. Scribes used many of the same conventions in their native tongues. If the VMS were Greek or Italian, for example, many of the same abbreviation conventions would apply except that the symbols are expanded into letters appropriate to that language.
So... AFA computat'nal attacks on lett'r frequency, et al, r concern'd...even if the VMS were a substitution code, the odds of it being one-to-one substitution are not very good. The use of abbreviations was deeply ingrained in the thinking of medieval scribes and the VMS has many short word-tokens (more than some of the transcriptions indicate).
RE: Computational Attacks on Abbreviated Text - ReneZ - 16-08-2017
This text has several deviations (that look like mistakes) from the standard text.
An obvious one is the sentence that should read:
Quote:Pene vero, quod est coniunctio, per E; poena, quod est supplicium, per OE
The word "poena" is written as "pena", which is in conflict with the point.
RE: Computational Attacks on Abbreviated Text - Koen G - 16-08-2017
I should probably point out that it's O'Donovan, not O'Donell
RE: Computational Attacks on Abbreviated Text - -JKP- - 16-08-2017
Koen, oops, my only excuse is fatigue. Thank you. I will fix it.
RE: Computational Attacks on Abbreviated Text - Helmut Winkler - 16-08-2017
JKP
If I dont misunderstand something he has written prepositio. But you are right of course that statistics are more or less worthles
RE: Computational Attacks on Abbreviated Text - Ruby Novacna - 16-08-2017
[color=rgba(0, 0, 0, 0.870588)]This is why we should work the text page by page and manually[/color]
RE: Computational Attacks on Abbreviated Text - -JKP- - 16-08-2017
This is a pretty typical example of abbreviated text from the 15th century.*
The first apostrophe is quite conventional (standing in for "ri"). The last one (the number 9) is a very common abbreviation for "-us" or "-um" and is the same symbol that is used as a VMS glyph.
The second apostrophe is somewhat unconventional and yet is completely understandable within this context, and illustrates the unpredictability of abbreviations, especially those in the middle of the text. As long as the context made the meaning of the abbreviation clear, scribes were at liberty to improvise and many did, sometimes using several abbreviations for the same word. Note that the apostrophe symbol (which is sometimes a line, sometimes a squiggle, sometimes a tail (an attached line), and sometimes an arc) is not always over the place where the letters have been omitted.
[*In case you're not familiar with Gothic cursive text, the word is Capricornus.]
There is a certain repetitious consistency to the VMS text that is dissimilar to manuscripts where abbreviations are used in flexible ways. There is also much less variability in VMS vord structure compared to natural language or even abbreviated natural language.
But... I've noticed that many of the researchers attempting to expand the text into Latin (and other languages) have assumed that some of the glyphs might stand for a variety of prefixes and suffixes (I made the same assumption when I was trying to make some sense of the text). The one that is most often expanded in different ways is the "y" (or "dy") that occurs with great frequency at the ends of words, probably because that is how the "9" glylph is handled in Latin and other languages using the Latin abbreviations (some of which were inherited from Greek conventions).
The "9" is not the only abbreviation glyph-shape used in the VMS. Many of these abbreviations were originally based on numbers, in pre-Carolingian and Carolingian days. The number 2 was also frequently used, most often representing "-ur" or "-tur". Sometimes the shape is tipped and resembles an r or c with a tail. A c with a tail can also be used for "cum" in Latin and "con" in Spanish.
One thing I'm certain of is that the VMS scribe was familiar with the glyphs used for Latin abbreviations. There are numerous examples and even a couple of places in which the 9 is superscripted. Here's an example from f33v, in which an apostrophe has been inserted to create dy'dy. Was this done to create a less common vord, or was the apostrophe added out of habit?
RE: Computational Attacks on Abbreviated Text - Koen G - 16-08-2017
Wouldn't that also imply that the scribe was 'thinking sounds' while writing? That he somehow assigned sounds to the glyphs?
RE: Computational Attacks on Abbreviated Text - Emma May Smith - 16-08-2017
(16-08-2017, 01:00 PM)Koen Gh. Wrote: You are not allowed to view links. Register or Login to view.Wouldn't that also imply that the scribe was 'thinking sounds' while writing? That he somehow assigned sounds to the glyphs?
I have nothing much to add, except that this is a great psychological insight. It's easy to explain the slip from a linguistic point of view, but not from a cryptological one.
RE: Computational Attacks on Abbreviated Text - Paris - 16-08-2017
(16-08-2017, 01:52 AM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.[*I'm not sure why the scribe has written this as Pre perpositio as I'm fairly sure Prepositio/Praepositio was intended.]
I'm also sure that the letter "p" with a tilde (~) means "Pre"
So, I read too "Prepositio".
You can check in this dictionnary of medieval abreviations which contains 800 abreviations used between 12th and 18th century.
You are not allowed to view links. Register or Login to view.