The Voynich Ninja

Full Version: Matching Plant Images Internally
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
(13-11-2025, 01:07 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.Why should I pay attention to Barchius' writing? He lived 200 years after the VM was written. He was even more clueless than we are today.

Because he had the VMS on hand for decades, and had hundreds of books on medicine and herbalism and hundreds of living doctors and herbalists that he could consult without wading through Google Ads and Sponsored Listings.  So if he says that the plants depicted in the VMS "have escaped observation here in Germany", it is strong evidence that the people who knew all about those "12000 plants" did not recognize them.

All the best, --stolfi
Yes, yes, and there were people who had a crystal ball in their hands for decades. They saw nothing, but were burned anyway. Big Grin

End of story.
I took the most convincing matches to see where they came from

[attachment=12374]

The best 8 pretty much unambiguous matches with scores of >2.5 are hands 1 or 5 and belong to the early quires with 3xQ3 (2 on same bifolio), 3xQ6 (2 on same bifolio) and one match in Q1 and 5. All these best matches are found on Pharma page foldout f102 (Q19) or on f89v2 (Q15)

Looking at all matches that have a score of at least 2 and were rated by more than one person, we get matches from all botanical quires 1-8, 15+17 with 6 matches from Q15 and two from Q17. There is no quire that does not have at least one good match in the pharma section.

From those best 32 matches:
24 or 75% are Hand1
5 or 16% are Hand2
3 or 10% are Hand5

Considering there are only 6 Hand5 plants, at least 50% are covered in Pharma.
In contrast, only 5 of 20 or 25% of Hand2 plants appear in Pharma though there are many prominent ones.
Curiously I estimate also approximately 25% of existing Hand1 plants are found in Pharma.

There is no strong match for a Hand3 plant though I would argue this is a somewhat decent match for f94r.

[attachment=12371]

It's also interesting that most of the best matches are clustered on one foldout page which has matches with plants from scribal hands 1,2 and 5.

[attachment=12372]

We also have some duplicates. You are not allowed to view links. Register or Login to view. , You are not allowed to view links. Register or Login to view. , You are not allowed to view links. Register or Login to view. , You are not allowed to view links. Register or Login to view. are depicted twice each. None are perfect matches though

As always, I have no idea what that means. I'll leave it to our statistics experts to determine if the Pharma duplicates are random samples of the botanical plants or not.

If we assume Hand1 is the earliest and Hand3 the latest, it makes sense that Hand3 plants and most Hand2 plants were created later. But why are so many of the rather poorly drawn Hand5 plants in Pharma?
(14-11-2025, 09:17 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.I took the most convincing matches to see where they came from

Have you tried to match words of pharma vs. herbal folios where the plants seem to match?
(14-11-2025, 09:17 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.I took the most convincing matches to see where they came from

Fantastic, tanks!

Quote:It's also interesting that most of the best matches are clustered on one foldout page which has matches with plants from scribal hands 1,2 and 5. ... why are so many of the rather poorly drawn Hand5 plants in Pharma?

Hands may or may not be relevant.  The Herbal may have been scribed in installments, over a long period of time, and a scribe presumably was recruited anew for each installment. And one of those scribes, or a different one, was recruited for Pharma.  Then, whichever of the two sections was done first, the process of copying from one to the other (or copying both from the same sketches) could have been independent of the scribes.

More interesting would be to compare Herbal-A and Herbal-B in this regard.  The "language" differences must be connected more to the Author rather than the Scribe.  Like, he changed the spelling rules, or copied from two different source books, or recorded two slightly different pronunciations of the same language...

All the best, --stolfi
(15-11-2025, 07:44 PM)quimqu Wrote: You are not allowed to view links. Register or Login to view.Have you tried to match words of pharma vs. herbal folios where the plants seem to match?
I'll leave text analysis to others, it should be simple to write a script that compares words, word variants and phrases in different pages - but not for me. As far as I know no convincing parallels were found between botanical and pharma pages but I might be wrong. I stick to imagery.

(15-11-2025, 09:53 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Hands may or may not be relevant. 
They probably are in regard to plant drawings, see Koen's brilliant work on Alpha vs. Beta plants.
You are not allowed to view links. Register or Login to view.

Currier A / Hand 1 plants are fundamentally different from Currier B Hand 2/3 plants. Hand 5 is somewhat intermediate. But yeah, the strongest distinction is between Currier A and B. Distinguishing Hand 2 and 3 is very hard though but the drawings have a different 'feel'

I have to somewhat revise my statement about Hand 3 plants being absent.
If we look at the flat-top root plant on f102r1 and acknowledge that You are not allowed to view links. Register or Login to view. by Hand2 is a decent match, we must also grant the same status to f95r2 by Hand3 which is highly similar. 


[attachment=12397]

I am however not totally convinced that these depict the same plant as in pharma. It has single leaves and this fish-tail flat top root. It also isn't a classic root table. The attached leaves form 'knubs' where they connect to the flat top of the fish-tail root. They are closed by a line which is an 'Alpha' feature. The corresponding 'Beta' plants by Hand2 and 3 have open attachments to the root table which is more like a dug out patch of soil, or some sort of 'claws' like small roots, a 'Beta' feature. This shows that while the Hand1 artist was aware of at least a prototype of the concept of root tables (which I believe is a misrepresentation of a patch of soil or a waterbody in the source image), he still drew the plant in his own stye which is different from the true root tables used only in 'Beta' plants by Hand 2/3, both of those hands are nearly indistinguishable. The small sample size doesn't help either.

This leads us to the fundamental question - was the Pharma section created before the 'Beta' style arose and the full plants corresponding to the Pharma sketches were drawn? At least I can't find any evidence that rules this out. The fundamental concepts were already established. The only thing we do not find in Pharma are VM 'daisies' but flowers are generally rare in Pharma.

The main points are:
All Pharma pages have at least some weak match to VM plants in the botanical section though the best matches are strongly clustered.
Multiple plants may show similarities to a single pharma image and vice versa
Plants of all scribal hands, both Currier A and B are found in Pharma with very good matches.
Plats from all botanical quires are found on Pharma with at least one decent match, mostly several.

I still wonder why the same plants were drawn multiple times on the botanical section, often but not exclusively by different hands. That's not overly unusual and is found in other manuscripts where the same imagery was re-used either because of an error or because no other image was available.
(14-11-2025, 09:17 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.I took the most convincing matches to see where they came from

Plants f1v, f18v, f23r, f32v, f37v, f47v are definitely matches: not just the same plant, but the "same" drawing.  Notably they match the root and leaves, but the Pharma versions omit the flowers.

Plants You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view. do not match at all.  Are they typos? Or maybe the match is on a different Pharma page?

Plant You are not allowed to view links. Register or Login to view. is a very poor match.  The only resemblance are the leaves, but that could be a coincidence.

Ditto for f94r.

Plant f57r, on the other hand, matches the flowers but not anything else.  This may be a case of copying, but maybe it is just a generic flower that the scribe would draw from his head.

All the best, --stolfi
There is also You are not allowed to view links. Register or Login to view. that matches f102v1 (note verso nor recto), bottom middle.

It seems that all the real Herbal-Pharma matches are with f102, recto and verso.
Using the notation -{row}.{column} for the latter:
  • f1v    f102r1-3.2  root, leaves. NO flower.
  • f18v  f102r2-3.1  root. NO leaves, flower.
  • f19r  f102v-2.3    root, leaves (type 1). NO leaves (type 2), flowers.
  • f23r  f102r2-3.2   root. NO leaves, flowers.
  • f32v f102r2-1.2   root, leaves, flower.
  • f37v f102r1-3.1   root, leaves. NO flower
  • f47v f102r2-1.1   root, leaves. NO flower.
Of the various possible explanations for these matches, I still lean to the Author having sketched only the important parts of many plants.  Then those sketches were copied to vellum by a scribe as the figures of the Pharma section.  Then other scribes created the Herbal pages, copying some parts of each plant either from a Pharma figure or from the corresponding Author sketch, and completing the rest of the plant either from their imagination, from nature, or from other herbals.

Makes sense?

EDIT: added f37v.

All the best, --stolfi
Page with clips of the above, side by side:
You are not allowed to view links. Register or Login to view.

Again, those are the real matches (not just similarities) between Herbal and Pharma that I know of.  They have score 3 in @Bernd's table.  One feature is that the Pharma sketch is always a proper subset of the Herbal figure; that is, every part of the former is in the latter, with same pose and perspective. 

Other entries of @Bernd's table, as far as I can see, are just resemblances that can be accidental.

That page doesn't include yet matches of Herbal and Herbal, like You are not allowed to view links. Register or Login to view. and f95r2.  

Did I miss any Herbal-Pharma?  

All the best, --stolfi
I didn't create this list, I took it from the file from the first page of this thread. Maybe you want to contribute to it? It seems I can't edit it though.
You are not allowed to view links. Register or Login to view.

You can read about the reasoning for each rating there in the sheet 'comments'. Of course it is quite arbitrary and as you can see different people have very different opinions on the quality and quantity of matches. I jut took all with a score of at least 2 that were rated by more than one person.
You are not allowed to view links. Register or Login to view. in the image is indeed an error, sorry about that. It's not in the list.
I just realized I have the above internal match You are not allowed to view links. Register or Login to view. - f95r2 in my list that should have been removed as it is not in Pharma. So we have one Hand2 plant less.

Your list is far more stringent, it's hard to decide where to draw the line.But I wonder why you would exclude those 2 Hand5 plants:
You are not allowed to view links. Register or Login to view. vs f89v2
You are not allowed to view links. Register or Login to view.
You are not allowed to view links. Register or Login to view. vs f89v2
You are not allowed to view links. Register or Login to view.

[attachment=12412]
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21