The Voynich Ninja

I went to the library and checked The Palaeography of Gothic Manuscript Books: From the Twelfth to the Early Sixteenth Century by Albert Derolez, Cambridge University Press, 2003

It is an excellent book, illustrated with clear diagrams of single character shapes as well as more than one hundred plates of manuscript photographs.
I found Derolez' discussion of the Cancelleresca subtype particularly interesting:

Quote:Since Italy was without doubt the cradle of Cursiva, it is no wonder that it was the first country to develop a luxury book script based upon Italian cursive chancery script. The so-called 'minuscola cancelleresca italiana' was used on a large scale from the end of the thirteenth till the end of the fourteenth century [...]. Kirchner called it 'Florentiner Bastarda'. [...] It has all the characteristics of Italian Cursiva Formata, but its style is particularly marked by its long looped ascenders ad long pointed descenders, both strictly vertical, and its small size of the letter bodies in comparison with the ascenders and descenders. The uncompressed character of the script is also counterbalanced by numerous fusions.
The generally triangular and highly conspicuous loops on ascenders consist of a bold diagonal downstroke and a closing hairline which is often invisible (if traced at all). [...]
g may have various forms, but never has the 'horns' present in Northern Cursiva. Scribes had a particular liking for g with a looped tail. [...]

I think that an important point is that the absence of 'horns' on g is typical of Southern Cursiva (even if there are no hard rules, of course). It seems clear that Cancelleresca has both similarities and differences with what can be observed in the VMS Marginalia. 
Anyway, among Derolez' plates, I think the closest to VMS marginalia is BAV Ott.Lat.1738 written in Padua in 1386. I was surprised by the early date. Derolez describes the script as "Italian Cursiva Libraria/Currens, the letter forms of which remind us of Cancelleresca". 
[attachment=2238]

Some letters are rather close (b,h,l,m,n,x), but the ascenders tend to curve forward and the loops are mostly less triangular than in the VMS.
d and final -s are markedly different. Long s and f are thicker, as one can often see in French manuscripts, but in this case they are at least vertical, with no slant. In general, the VMS script tends to be simpler, with less shading and flourishes. 
But I think that this in an interesting parallel and I hope that exploring Veneto sources of the time could provide other interesting results.

Great work, Marco! 1386 is surprisingly early indeed for such a close match, though on the other hand it's "only" two decades before the VM date range. Still, I'd been expecting matches later rather than earlier.

Does Derolez explain if the term cancelleresca is just a name, or was it really mostly an administrative script?

Excellent Marco!

The strange thing is that, while the individual characters are indeed an excellent match, better than any I found, the "look and feel" of the text as a whole is quite different from the Voynich MS.

The most important conclusion I see is that one cannot say at all that the hand of You are not allowed to view links. Register or Login to view. is a German hand. Even though it clearly includes German words.

(05-07-2018, 09:35 PM)Koen Gh. Wrote: You are not allowed to view links. Register or Login to view.Does Derolez explain if the term cancelleresca is just a name, or was it really mostly an administrative script?

Hi Koen,
Derolez's book is too small and with too wide a scope to enter into the details. He says Cancelleresca is derived from Italian cursive chancery script. He also says it was mostly used for vernacular books, in particular several copies of the Divine Comedy are written in this style: he provides an example in Plate 89 (left): Laurenziana, Gaddi 90, Florence 1347. The other example he provides (Plate 94, right) is a Latin work by Statius (Beinecke, Marston 42, Italy, 1406).
From what I have read on various online pages, the script was created in Florence and chanceries of the various Italian states had their own variants. These scripts were rather complex, since one of their goals what to prevent forgeries (like the Imperial Elongatae we discussed elsewhere). The book version discussed by Derolez is not so complex and the Latin script in the VMS even less so.

[attachment=2240]

(06-07-2018, 12:49 PM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.The strange thing is that, while the individual characters are indeed an excellent match, better than any I found, the "look and feel" of the text as a whole is quite different from the Voynich MS.

Hi Rene,
I agree on the look being different. The execution of the VMS Latin script is clearly of an inferior quality in comparison. Also, BAV Ott.Lat.1738 makes ample use of hair-traits for abbreviations and flourishes. I hope it will be possible to find other candidates written by less skilled copyists than the author of the BAV manuscript.

(06-07-2018, 12:49 PM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.The most important conclusion I see is that one cannot say at all that the hand of You are not allowed to view links. Register or Login to view. is a German hand. Even though it clearly includes German words.

I don't think there are reasons to classify the hand as German, but You are not allowed to view links. Register or Login to view. mentioned by JKP also is an interesting (if late) parallel and it was likely written near Vienna. I guess this Cancelleresca style might have been adopted in some Southern German-speaking regions? We will see what else can be found....

I like to let the math tell me how similar a script is to the VMS text.


MarcoP, thanks for uploading the sample for Ott.lat.1738. I added it to the other samples and got a score of 67 out of 120.

Out of a total of 1,000 plus samples chosen for their similarity to the marginalia, I have 205 that score 67 or better. Here are a few examples.

The VMS marginalia is at the top, Ott.lat.1738 at the bottom. Note how similar Ott.lat.1738 is to the one that is two slots above it (which is thought to be from Bavaria), a few characters are different (especially the r and v), but there are also a few similarities (to this and to the other samples):



The VMS character at the end is not actually a "z", it's an abbreviation symbol that resembles "z", but I store it in that slot for reference.

The long serif on the VMS "i" is quite distinctive, as is the high curve on the stem of the "t" (possibly a throwback to the older form that was curved like a "c").

I also have some information on the letter "a" You are not allowed to view links. Register or Login to view. (for some reason these aren't showing up on the blogosphere).

You are not allowed to view links. Register or Login to view. is a German manuscript, 1400 ca. The script is a rather messy Cursiva, with two-storey a that by itself makes it different from the VMS Latin script. But there are other character shapes that are comparable with those in the VMS, e.g. p tends to have an open top and x a right-side loop. I think the most interesting feature is the 8-shaped and high final s: considering the variability of the script, the 8 seems rather regular.

The Latin text is available You are not allowed to view links. Register or Login to view..
The attached detail reads:
importat omnes homines singulariter ab=
solvendi a culpis et penis sub specie pie=
tatis caritatis et misericordie iam dicto titulo
plene potestatis per Romanos episcopos sibi primum
assumpto; ex hiis paulatim atque latenter

It would be interesting to see if a similar 8-shaped final S also occurs in manuscripts from different areas.

The 8-shaped final-s is not very common. I haven't found very many. It took a lot of effort to locate a couple of dozen somewhat close matches, and most of them are fatter than the VMS figure-8 (which researchers variously identify as "s" or "d").

I took a look at Ms. 28. Sometimes the final-s is B-shaped, sometimes almost figure-8 shaped. I grabbed a section where there were several of them because I was curious about the stroke order:





-------------------
In most manuscripts, the final-s is B-shaped or shaped like Greek sigma. Here are some examples from scripts with final-s that don't match the VMS figure-8, but which are similar to the VMS marginalia style in other respects:





And here are some samples of final-s letters that are closer to the figure-8 style. These are specifically used at the ends of words (with straight-s or long-s in other parts of the word). The VMS figure-8 character is not limited to the final position, so it's not 100% certain it is a final-s or even an "s" but it appears to behave like "s" in a number of ways.

The date range for the closer matches came out as c. mid-14th century to late 15th century. The origin of many of the manuscripts below is unknown. Known locations and numerous guesses by historians for this small selection include Alsace, Bohemia, Belgium, Provençe, Flanders, England, Moravia, Italy, France, and many different locations in Germany:




It took years to find these, and yet I don't think any of them are super close. The angles of the loops in the VMS figure-8 are distinctive. The bottom loop of the two VMS glyphs on the left are almost rhombus-shaped.

[font=Eva][font=Eva][font=Tahoma, sans-serif]JKP
[/font][/font][/font]
In the table on your site there are no examples of the letter "a" in the form of "[font=Eva]a". Does this mean that "a" is typical only for VMS?[/font]

How to explain the presence of the symbol "[font=Eva]a" in the word "ALADABAS" on 116v?[/font]

1 / chance.

2 / the author of the marginalia was involved in writing VMS, but accidentally forgot to switch the writing style of the character "a".

3 / the author of the marginalia was involved in writing VMS, (or not involved, but understood the language of VMS). He deliberately wrote the syllable "[font=Eva]ad", because the last line also uses VMS characters.[/font]
[font=Eva][font=Eva][font=Tahoma, sans-serif]4/ ? ....  [/font][/font][/font]

(07-07-2018, 05:49 AM)Wladimir D Wrote: You are not allowed to view links. Register or Login to view.[font=Eva][font=Eva][font=Tahoma, sans-serif]JKP
[/font][/font][/font]
In the table on your site there are no examples of the letter "a" in the form of "[font=Eva]a". Does this mean that "a" is typical only for VMS?[/font]

I'm sorry, the blog post is a bit long, it's easy to miss the table. You can find a table of one-story marginalia "a" in the middle, after it says, "Note that the stem is fairly vertical, and...". It's combined with a couple of other letters, which may have confused things. I haven't posted matches to any of the regular text yet. I'm still organizing it.

Quote:How to explain the presence of the symbol "[font=Eva]a" in the word "ALADABAS" on 116v?[/font]

1 / chance.

2 / the author of the marginalia was involved in writing VMS, but accidentally forgot to switch the writing style of the character "a".

3 / the author of the marginalia was involved in writing VMS, (or not involved, but understood the language of VMS). He deliberately wrote the syllable "[font=Eva]ad", because the last line also uses VMS characters.[/font]
[font=Eva][font=Eva][font=Tahoma, sans-serif]4/ ? ....  [/font][/font][/font]

The "a" in the marginalia is very different from the "a" in the main text. It doesn't seem like the marginalia writer was one of the main-text scribes, but so many of the letters in the main text have that deliberate back-leaning slant (including the stem on the "a"), it's hard to tell if the letters were deliberately written differently and thus might look considerably different from the scribes' regular writing.


I have to admit though, because of the way the "g" is written in the marginalia (it's quite elegant), I sometimes wonder if the marginalia writer was artistic and maybe contributed to the drawings, or maybe was attracted to the manuscript because of the drawings.

(07-07-2018, 06:05 AM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.[quote='Wladimir D' pid='21343' dateline='1530938980']
[font=Eva][font=Eva][font=Tahoma, sans-serif]JKP
[/font][/font][/font]
In the table on your site there are no examples of the letter "a" in the form of "[font=Eva]a". Does this mean that "a" is typical only for VMS?[/font]

I'm sorry, the blog post is a bit long, it's easy to miss the table. You can find a table of one-story marginalia "a" in the middle, after it says, "Note that the stem is fairly vertical, and...". It's combined with a couple of other letters, which may have confused things. I haven't posted matches to any of the regular text yet. I'm still organizing it.

Quote:How to explain the presence of the symbol "[font=Eva]a" in the word "ALADABAS" on 116v?[/font]

1 / chance.

2 / the author of the marginalia was involved in writing VMS, but accidentally forgot to switch the writing style of the character "a".

3 / the author of the marginalia was involved in writing VMS, (or not involved, but understood the language of VMS). He deliberately wrote the syllable "[font=Eva]ad", because the last line also uses VMS characters.[/font]
[font=Eva][font=Eva][font=Tahoma, sans-serif]4/ ? ....  [/font][/font][/font]

The "a" in the marginalia is very different from the "a" in the main text. It doesn't seem like the marginalia writer was one of the main-text scribes, but so many of the letters in the main text have that deliberate back-leaning slant (including the stem on the "a"), it's hard to tell if the letters were deliberately written differently and thus might look considerably different from the scribes' regular writing.


I have to admit though, because of the way the "g" is written in the marginalia (it's quite elegant), I sometimes wonder if the marginalia writer was artistic and maybe contributed to the drawings, or maybe was attracted to the manuscript because of the drawings.

--------------------
Edit [addition]: Wladimir, I started a thread on this You are not allowed to view links. Register or Login to view. as I think you bring up an interesting point.