The Voynich Ninja

(28-11-2025, 08:32 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.I had to remove a line from the article about the mineral You are not allowed to view links. Register or Login to view. that claimed that it was used as a paint pigment in the VMS.  Just because the McCrone technician copy-pasted the output of their spectrum-matching software onto the report, without checking what palmerite was.

I just removed that reference to the VMS from the You are not allowed to view links. Register or Login to view. article.  And another one from the You are not allowed to view links. Register or Login to view. article.  See the respective Talk pages.  Please let me know if you find other cases.

To be honest, from where I stand the removal of the plant ID hypotheses is fully justified, but the removal of the line about palmierite is doing the opposite: replacing information from a technical report by an opinion of a non-expert. 
Strictly speaking, the report says: "possibly minor amounts of lead sulfide and palmierite", so at least the word "possibly" (or equivalent) should have been there.

However, I don't want to make a point of it, because the whole article says noting at all, and having just the Voynich MS reference doesn't really make any sense.


In earlier days there was a similar discussion about atacamite, which some people thought was non-European. This material was found in other pieces of art in Italy, though. (This is from memory so "C.E.").

(28-11-2025, 09:05 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.replacing information from a technical report by an opinion of a non-expert. 

My opinion is based on the data included in the report.  I understand enough of the physics and chemistry involved to see how they got that palmierite diagnostic.  The report includes a noisy X-ray diffraction spectrum and the output of a software that looks for the best match to the spectrum in a large library of spectra of minerals.  The software said "palmierite", and the technician simply copy-pasted that into the table at the end of the report.  He obviously did not even bother to check the Wikpedia article about that mineral.  If he had, he surely would have felt the need to explain how that exotic mineral, found only around fumaroles in a handful of places in the world, could have found its way in that paint.

(28-11-2025, 09:05 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.the whole article says noting at all

The reference in that article has a bit more information.

(28-11-2025, 09:05 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.In earlier days there was a similar discussion about atacamite.

Natural atacamite was discovered in South America, but there are medieval recipes for creating a green copper pigment that turns out to be (synthetic) atacamite.  

So at least atacamite is something that could have been used in a paint at the likely dates when the VMS was painted -- which means any date before Voynich got it.  

However, I personally don't think that is likely in the case of the VMS, specifically, since the recipe for synthetic atacamite is not trivial (expose a plate of copper covered in honey to vapors from vinegar for months), and the Painter seems to have been very amateurish.  It seems that he did not even have a decent brush.  I can't see him making his own pigments according to obscure alchemical recipes.

Besides, the Painter already had a green pigment, which he used liberally all over the VMS. The McCrone technician could not identify this pigment beyond saying that it contained copper but was not crystalline; hence he guessed "a copper resinate", which does not mean much actually.

All the best, --stolfi

I looked up palmierite and this is one of the first links: You are not allowed to view links. Register or Login to view.

How does a random reference to the Voynich end up there? Probably taken from the Wiki... 
Even if actual palmierite were found in the VM, it's so random and confusing to introduce that into a three-sentence stub about the mineral.

Stolfi: I see what you're saying here, and I agree that we must think of this as the output of software rather than something proclaimed by a learned professor. However, I think the writer was still correct in including this sentence. If the reading is consistent with something that is possibly palmierite, then that information should not be omitted just because the "possibly" is a more like a "probably not". If it turns out that the VM was made with materials from one of these rare palmierite places, then having omitted this possibility from the report would be seen as negligent.

I wonder with minerals like these, do they have to come from a natural deposit? Or can trace amounts be explained by some chemical processes taking place during ink preparation, with contaminants etc?

(28-11-2025, 11:41 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.and I agree that we must think of this as the output of software rather than something proclaimed by a learned professor.

Just for the record, and without any animosity, I disagree with this.

All this was ages before AI, and all was done by humans with very specific knowledge, interpreting and evaluating data. The analysis of the Voynich MS was quite a high-profile activity, and not the only high-profile activity of McCrone.

If another materials scientist were to challenge the results, then I would listen.
In the present scenario - not.

If the report says: "possibly minor amounts of lead sulfide and palmierite", and the amateur response is: no, definitely not, then that does not look like a winning argument to me.

(28-11-2025, 11:57 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.If the report says: "possibly minor amounts of lead sulfide and palmierite", and the amateur response is: no, definitely not, then that does not look like a winning argument to me.

What I'm trying to say is that we don't have a scholarly paper on how likely it is to have palmierite in the VM. It's not like professor so and so proclaimed this after weighing all the data and the historical context. There is no assessment of the likelihood of the palmierite scenario, because that's not expected of the report. This is more likely the result of a (manual or automated) comparing of test results to reference data.

What I'm saying is that even if an automated process determined that the reading could point to palmierite, then it should still be in the report and we have to take it for what it is.

(28-11-2025, 11:41 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.I wonder with minerals like these, do they have to come from a natural deposit? Or can trace amounts be explained by some chemical processes taking place during ink preparation, with contaminants etc?

It's not my area, but recently there have been a couple of cases of it popping up in scans of paintings. 
Some reports claim this is due to the environment the work was kept in at some stage of it's life (not part of the paint). 

I should also say that I have no idea at all if "a mixture of palmierite (K2Pb(SO4)2) and anglesite (PbSO4)" makes this an entirely different thing or not, or if its ok to just connect palmierite to palmierite. 

You are not allowed to view links. Register or Login to view.

This is what the news article is referencing I believe - You are not allowed to view links. Register or Login to view.

(28-11-2025, 11:41 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.How does a random reference to the Voynich end up there? Probably taken from the Wiki..

Yes, that is one of many sites that repackage Wikipedia articles.  "LLMs before LLMs"...

(28-11-2025, 11:41 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.I wonder with minerals like these, do they have to come from a natural deposit?

The Wikipedia article lists a reference that gives a little more information:

• "Palmierite is a rare fumarolic sublimate occuring at a handful of localites worldwide. It is a lead sulfate. According to the old Cureton tag this is from the eruption in 1924. The yellow [in the photo] is perhaps some sulfur or one of the many sulfates such as Erythrosiderite or Metavoltine. The Palmierite are tiny colorless plates as is the Aphitalite. Vesuvius complex contains an incredible 39 type locality minerals, this being one of them."
  

Thus, apart from the extreme rarity, inaccessibility, and recent date of discovery, no one would think of using palmierite as a pigment.

(28-11-2025, 11:41 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.can trace amounts be explained by some chemical processes taking place during ink preparation, with contaminants etc?

The chemical formula is very simple and I would not be surprised if traces of it it can form if the elements are present.  But just saying "palmierite detected" without discussing this point is by itself quite misleading.  We can see the confusion that it caused, including that spurious reference in Wikipedia.

And, again, the "detection" is very likely to be just a false match.  

All the best, --stolfi

(28-11-2025, 11:57 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.If another materials scientist were to challenge the results, then I would listen. In the present scenario - not.

Rene, what can I say? Obviously your a priori confidence in experts is much higher than mine.  Perhaps because, in your profession, experts must be really experts?  In mine, unfortunately, and in many others, I have seen many "experts" who get by only because the people who use their opinions cannot tell when they are just guessing, or flat out wrong.  

Or even intentionally lying in order to protect their own interests, like J. Eric Thompson about Maya writing 

And then there is the academic "omertà" -- "I will not speak out about your incompetence if you do not speak out about mine."

And then there is the problem that an expert's opinion about X is not very significant, if his expertise was acquired on zillions of cases which are all different from X.   Like the VMS is different from all the zillion manuscripts that experts are familiar with.

When I was a teenager, at some point I had a dozen fish tanks and bred many kinds of fish, from plain guppies to bettas and such. I read aquarium manuals and knew by heart the binomial names of all the fishes, plants, snails, worms etc in my tanks.  Since then I watched hundreds of documentaries of fish, including many of Jacques Cousteau's and The Life Aquatic with Steve Zissou, and learned all about fishes that fly and fishes that walk, sea horses and sea basses, coelacanths and molas; and I bought and cooked lots of fish, and ate lots of sushi and sashimi.   

If that does not make me an expert on fishes, I don't know what would.  And one thing that I learned from all that experience is that all fish have gills.   Therefore, when I tell you that whales have gills, you'd better believe it.



All the best, sincerely, --stolfi

(28-11-2025, 12:16 PM)Koen G Wrote: You are not allowed to view links. Register or Login to view.There is no assessment of the likelihood of the palmierite scenario, because that's not expected of the report.

And, by the way, hey have one X-ray instrument that can tell how much of each element is there in the sample, but not in which chemical form.  

The instrument that detected palmierite is a different one, that directs a fine beam of X-rays onto a microscopic particle taken from the sample, and measures how much of the beam is scattered in which directions as the particle is turned.  If the particle contains a single crystalline substance, that data is a spectrum (like that on page 2 of the report) whose peaks depend on the spacing and arrangement of atoms in the crystals of that substance. The software then scans a large library of spectra of minerals and reports the best match (azurite in the case of page 2).  

But that only identifies the crystalline substance in that particle that was analyzed. It cannot tell how much of that substance is present in the paint.  Probably not even in that particle.  And it cannot identify non-crystalline substances (like iron-gall ink, or the green pigment most used in the VMS).  And there may be several spectra in the library that are equally good matches to the data, within the limits of the noise; but the software apparently prints only one...

All the best, --stolfi

(28-11-2025, 05:46 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Rene, what can I say? Obviously your a priori confidence in experts is much higher than mine. 

Higher, yes, probably. Generally? Yes, also probably.

I just cannot agree with the complete lack of confidence, and the assumption of amateurish behaviour (as described), in this particular case.
 
This is a discipline exercised by different centres, who make publications, advance their knowledge, exchange knowledge at conferences etc. How can we judge, even have the first idea, about the extent of knowledge these people have?