The Voynich Ninja

With MSI (Multispectral Imaging) it was possible to recover faded text (example: 1r), but the scans also capture parts from the facing page (example: 26r shows offset from 25v). In case there are ever MSI scans of the whole book (or at least of the pages facing the missing ones - which could possibly be the case already), we could get a glimpse of what was on those missing pages.

I don't know much about post-processing, but one would have to subtract the normal scan (and the show-through) from the MSI scan. Obviously this would only reveal very little, but that shouldn't stop anyone from trying.



Has something like that done on any other manuscript?

It would be possible to fine-tune this process with the scans we already have, since we know what the process is supposed to reveal.

Additionally this could be of great help to restore the original order of the pages!

Hi!

As far as I remember, Lisa Fagin Davis said that this would be great if some information about the missing pages was retrieved this way, but personally I'm not sure there is enough data there. Which offset on 26r are you referring to? If it's offset from the paint, then one may not even need MSIs, offsets from the paint are quite visible on many pages and probably some ML algorithms can identify them using the visible light scans. Have any offsets from the text or line drawings been detected in the MSIs? If the book was bound long after it had been written, which seems likely, there might be no offsets from the ink at all.

Indeed, I do think it is possible that offsets from the original sequence of leaves might be visible under MSI, although there's no way to know for sure what might be found until and unless it happens. Unfortunately, I have been told by several different decision-makers at the library that it isn't going to happen, at least not anytime soon. Lots of reasons - staff time, concerns about exposure and handling of the manuscript, and competing library priorities. I'm hopeful it will be possible someday. Even if we did have a complete set of images, the offsets would likely only be from pigment, not ink. So while we might be able to infer what genre of pages were lost, it is very unlikely that MSI would reveal offsets of text.

(16-12-2025, 03:32 PM)LisaFaginDavis Wrote: You are not allowed to view links. Register or Login to view.Unfortunately, I have been told by several different decision-makers at the library that [full multi-spectral imaging] isn't going to happen, at least not anytime soon.

We all wish we had as many images as possible.  But I don't think that the information we could get from more MS images would be worth the cost and hassle of acquiring them.

We all had hoped that the MS images would reveal things we could not see in the BL 2014 ones.  Like text that was erased or faded to invisibility; or significant text hidden under the opaque paint.  But that largely did not happen, did it?

AFAIK, the significant new information we got from the MS images was

1. The ink is not iron-gall ink.
   
2. Most of page f116v was soaked in water.
   

Item 1 explains why so much of the text has faded so badly, and makes the VMS atypical in yet another way.  But otherwise does not seem to have much impact on anyone's research.

Item 2, in my view, solves the riddle of the text on You are not allowed to view links. Register or Login to view. -- although in a rather disappointing way.

What else have we learned from those MS images?

So, rather than more MS images, I would rather have more high-resolution, visible-light images of selected details.  Like this image from the McCrone report:
[attachment=13009]
Such images could help us prove or disprove two important open questions:

1. Is the painting original, or a much later addition?
   
2. Is all the ink original, or was some of it retraced or added later?
   

The answers to these questions can have significant impact on many people's research, e.g. attempts to identify plants or to deduce the origin of the Scribe by the decorative details he used.

Unlike the MS images -- which demand complicated setups, expensive equipment, and specialized personnel -- such high-resolution images could be acquired by any trusted visitor with a hand-held digital microscope, with no special setup and with no stress to the book.  If the Library would only allow it....

All the best, --stolfi

I believe folio 93r is next to a missing bifolio and is among the MSI scans, but if the bifolio was lost very soon after the manuscript's creation, I'm not hopeful for much of an offset. I'm assuming we would have a much better chance with the two that were cut out centuries later, 12 and 74. 12 would just be another couple plants from the section of the manuscript we already have the most of, but 74 would potentially give more zodiac illustrations to estimate the date and location of the manuscript with.