Voynich – the McCrone report now online!
Wowza – the long-awaited chemical analysis of the Voynich Manuscript’s inks by the McCrone Institute (you know, the one commissioned for Andreas Sulzer’s 2009 ORF documentary on the VMs) has just appeared sans fanfare on the Beinecke Rare Book & Manuscript Library’s Voynich page.
Feel free to read the report as a PDF, though note that it wouldn’t render in Internet Explorer for me, so I downloaded it directly (“Save Target As…”) and opened it in Adobe Reader. Its key conclusions are:-
* A single ink [typical iron gall] was “in all probability” used for both the main body of the text and for the drawings.
* A second ink [high iron] was used for the folio numbers.
* A third ink [high carbon, very low iron] was used for the quire numbers .
* A fourth ink [high carbon, very low iron] was used for the Latin alphabet on f1r.
* The blue paint was ground azurite “with minor amounts of cuprite, a copper oxide”.
* “The green paint is a mixture of copper-stained amorphous organic material optically consistent with copper resinate, and copper-chloride compounds consistent with atacamite or similar compounds”, but without any resins obviously present.
* Gum (presumably gum arabic) was used to bind the green paint and all the inks (apart from the Latin alphabet ‘a’ on f1r, which seems to have been bound with a protein), though “the spectra include several sharp peaks [...] that are not expected for a gum as per the spectra in our library”, which “suggests the possibility of other constituents, which remain unidentified as of this date”. Note that the blue and red-brown paints were not tested for gum.
It’s going to take a while to digest this properly, basically because the Beinecke has only released the text part of the report, and none of the figures, photographs or reference spectra mentioned in the text. Other scans referred to in the text (such as UV scans of f1r, and presumably of f17r as this appeared in publicity montages for the documentary) are similarly absent: it would be particularly nice to see these as well, wouldn’t you say?
To my mind, the various ink compositions would seem to suggest that there were three distinct codicological phases: a first text/drawing phase (normal iron gall), a second quire number and f1r Latin alphabet phase (where the inks are different, but made to broadly the same house style), and a third folio numbers phase. All of which should be no great surprise to most Voynich researchers, but all the same I personally find it interesting that the quire numbers seem to have been added in the same general phase as f1r’s attempted cipher alphabet. It therefore seems likely that the quire numberer did not know how to decipher the VMs, a conclusion I reached several years ago via quite independent codicological means.
Finally, it is somewhat disappointing that the single most-debated piece of information is conspicuously absent: I refer, of course, to the suggestion that the ink was added not hugely long after the vellum was originally made. Which unfortunately means that many of the nuttier theories are still in play. Oh well: apart from that, it’s a nice piece of work, highly recommended!
June 1st, 2011 at 9:56 pm
Thanks Nick… great news, thanks. I posted a link to your announcement on the VMS-net. Rich.
June 2nd, 2011 at 12:46 am
Very curious that the Beinecke has not taken that opportunity to alter the date it suggests for the manuscript. Have we some problem with the carbon-dating?
Beinecke para still reads “late fifteenth or early sixteenth century”.
June 2nd, 2011 at 12:47 am
[...] für die Dokumentation, die 2009 gesendet wurde) ist jetzt veröffentlicht — und wenn nicht Nick Pelling in seinem Blog darauf hingewiesen hätte, denn hätte ich es gar nicht mitbekommen. Das achtseitige Dokument steht auf der Voynich-Seite der [...]
June 2nd, 2011 at 5:27 am
Thanks, Nick (and Rich!).
Just a little observation to say that the folio numbers and the “michiton oladabas” text also seem to had been written with the same ink (high iron content), so it would be a part of your “third phase”.
June 2nd, 2011 at 11:40 am
Dear Nick, all,
as I have been trying to point out (quite unsuccessfully, it seems) in the past, the McCrone analysis does not say anything about how long after the vellum creation the ink was applied.
This is why this is not listed in the report.
There was a general statement made along those lines in the press conference (but not by McCrone!), which has been propagated into the newspaper summaries, largely to this effect.
Note that I was present at the press conference – I am not making this up.
Somehow I feel, however, that this piece of misinformation will not die out just yet….
Cheers, Rene
June 3rd, 2011 at 7:27 am
José: you’re right… but there’s a lot more still to be said about this, and it will require a more detailed follow-up blog post to do it all justice.
June 3rd, 2011 at 10:07 am
I’m new here. But I’d like to make some contribution to the Voynich MS. Has anyone thought of the voynichese as alliterative words? I mean alliteration was quite used in the Middle Ages all over Europe. Although English poetry used much more frequently. I’ve found Petrarca used alliterative lines in his poems.
So the repetitive nature of some words or some sounds in the voynichese lang.could be alliterations of some graphs. Well it was just an idea.
June 3rd, 2011 at 4:11 pm
I am most curious about the McCrone assertion that their finding of copper and zinc was “unusual”. If anyone has contact with the staff, please ask when, if ever, it was “usual” to find these elements in ink. Also, they suppose their presence could be caused by the ink coming from a brass inkwell… so is it known that brass inkwells can infuse ink with copper and zinc, or is this a guess on McCrone’s part? And lastly, can a brass pen nib either infuse ink with copper and zinc, or can copper and zinc rub off on a vellum/parchment surface from a nib, in normal use?
June 4th, 2011 at 7:46 am
Well I’m happy. Pretty much all the copying done in the same place, at much the same time (one supposes) and by people either in the oak-gall belt, or with acces to a trade in same (was there one?).
Wonder to what purpose, zinc was put in the fifteenth century?
June 4th, 2011 at 8:02 am
Paper on distribution of gall-producing wasps at
http://www.biomedcentral.com/1471-2148/10/322#
June 4th, 2011 at 8:44 am
I’m assuming the zinc was mineral, not organic. Nor it is clear when the spectra suggested the presence of pure zinc – or do I misread?
Anyway “Brass, which is an alloy of copper and zinc, has been used since at least the 10th century BC. Impure zinc metal was not produced in large scale until the 13th century in India, while the metal was unknown to Europe until the end of the 16th century.” The refs only wiki, but this might explain why Beinecke wants the date higher. Old story – Europeans want to think that no-one went from the Med to India before daGama. Twits.
June 4th, 2011 at 8:51 am
or we could be looking at the Vinland map makers.
“aluminum, zinc and gold that are present in the VindandMap ink strongly”
Nick – please feel free to edit/telescope these consecutive posts iif you wish
June 4th, 2011 at 9:19 am
It is interesting but not that big a surprise that tests on the black inks differentiate between page numbers, quire numbers, the marginalia on folio 1r and the actual Voynichese text. Some of us, including Nick, had guessed this from palaeographical considerations.
A significant negative result is that the tests do not bear out the hypothesis that there were two or more scribes. Sample 6 (47r) – said to represent Currier language A, scribe 1 – is not distinguished from samples 1, 9 and 13, said to represent language B and hands 2 and 3 (according to the notes in the EVMT transcription). Nor do the tests give us reason to think that sample 16, folio 116v, the pen trials, is distinct from the rest of the text. It may still be true that there was more than one scribe, but there is nothing in these results to substantiate the idea.
June 4th, 2011 at 9:54 am
Philip: all true, though I think it’s worth adding that the palaeographic complexities of f116v could never be resolved by a single ink sample (Sample 16) – the final chapter on the VMs’ marginalia has yet to be written. Also, I suspect the overall lesson about ink preference will turn out to be that it can be more of a long-term codicological constant than handwriting: I’ve suggested that Currier Hand 2 was no more than Currier Hand 1 but with smaller nibs (say, an eagle feather?), and that doesn’t seem to be obviously disproved by this class of evidence.
June 4th, 2011 at 10:57 am
er – and tutia of course.
June 4th, 2011 at 1:44 pm
Sorry Nick-
I should have consulted Nicholson and Shaw before getting in a tizz about zinc and atacamite.
Sure enough, the latter is a pigment found in analysis of colours used in papryi and zinc is a component of copper-bearing ores in the eastern desert. Large metal-workings and foundries etc at just the time and places that the imagery also indicates. For the later period, high carbon, low iron inks typical of Indian and Arab inks, while the usual pen-case (Qalander – spellings vary) is made of metal, and has inbuilt inkwell of the same material, so no probs there either. Cheers.
June 5th, 2011 at 2:46 am
Have already mentioned this on the list, but might be useful here Nick:
Mc Crone did an analysis on some Egyptian (common era, Coptic) texts, and here’s what they said about them:
“McCrone Associates, a firm specializing in forensic ink analysis, conducted a transmission electron microscopy (TEM) test on samples of the document’s ink. The procedure uncovered the components used to create the ancient ink and found that they are consistent with ingredients in known inks from the third and fourth centuries A.D. The ink includes a carbon black constituent, in the form of soot, bound with a gum adhesive. An additional procedure, Raman spectroscopy analysis, established that the ink also included a metal-gallic component like those used in third-century iron-gall inks. McCrone Associates reports that the Gospel of Judas may have been penned with an early form of iron-gall ink that included a small amount of carbon black (soot). If so, it could be a previously unknown “missing link” between the ancient world’s carbon-based inks and the iron-gall alternatives that became popular in medieval times.”
Cheers
June 7th, 2011 at 11:10 am
It’s about atacamite. A french natural history museum says – (this is good for your thesis, Nick):
“one finds atacamite in Italy, Gemany and in Greece.
In France it is present in the cupriferous regions of Cape the Garonne in the Var, in Daluis (Alpes Maritime), close to Gisoni (Corsica) as in the Bas-Rhin (lower Rhine), as in Triembach la Valee”
http://www.museum-mineral.com/recherche/?q=atacamite#http://www.museum-mineral.com/especes/atacamite,149.html
the bit about being used to dry inks came up in the web search, but I think needs a sign-in to find on the site.
June 7th, 2011 at 11:25 am
and Madagascar
http://www.answers.com/topic/atacamite
June 7th, 2011 at 11:44 am
Madagascar, eh? That’ll be handy if I ever have to screenwrite a Voynich-themed sequel to an animated children’s film. Which I’m not ruling out, BTW.
June 7th, 2011 at 11:47 am
Diane: thanks very much! It would be even more interesting if each of the various sources of atacamite you list has its own distinct chemical signature… As always, you never quite know to where any given VMs clue trail will lead (though Madagascar might possibly be stretching it a bit).
June 7th, 2011 at 1:29 pm
It does.
I refer to the relevant paper in my lastest blog post. Too annoyed with myself to add more:
http://anothervoynich.blogspot.com/2011/06/atacamite-again.html
June 7th, 2011 at 1:57 pm
… but of course you knew that.
Thanks Nick.
June 7th, 2011 at 2:12 pm
Diane: I should have added “…and the McCrone people have all the different reference spectra in their library”. The first problem is that it’s far from clear (from the all-too brief mention in the report) whether what was noted in the spectra was indeed atacamite, or one of the many very similar forms of the same thing, or (as you point out) a byproduct of paratacamite breaking down, etc. It’s not yet as solid an evidence class as we would like – I’ve asked the Beinecke if it will post scans of the figures in the report on its website, so hopefully these should help firm this up, even if only a little.
June 7th, 2011 at 2:36 pm
Nick if I were you, or someone else with likely contacts to the gentleman who pointed out which pages to analyse, I’d talk to him.
Someone has certainly had a full two years in which to use that information and write a fairly hefty tome on the Voynich.
I’ve decided not to go with my first response – whch was to scrap my blog and leave them to it I mean the Beinecke et. al.
Instead, I’ll get my hard copy of a post about minerals and fol.86v + rondels, and repost the dam thing. It might give others a bit of a chance, anyway. But I do hope my childish efforts to decipher the script – (I suggested the labels on those rondels were an effort to transliterate Theophrastus’ Greek using a version of the south arabian scripts) wasn’t correct. Those posts I took down after a few months, but what bothers me is that all the mineral comments are not to be seen – save the key (without minerals) which I posted as a separate page.
I’m not imagining anything sinister. Blogger recently had massive problems, and many posts temporarily vanished. Some,and numerous illustrations, just never reappeared.
There are many reports on Raman spectroscopic analyses on the web. Failing that, phone Frost. Decent chap I gather. Don’t know him though.
June 7th, 2011 at 3:18 pm
Diane: the McCrone analysis was sanctioned by the Beinecke curators, but was paid for by the documentary makers – so I suspect it was simply the case that they didn’t want to release the data until after it had eventually aired in the US. I doubt there’s anything more sinister than that going on here: I also strongly doubt there’s any kind of hefty Voynich tome in the offing.
June 8th, 2011 at 10:22 am
Raman spectroscopy will identify minerals, and paratacamite is separate mineral from atacamite. It was defined so in the fifties, but I understand it is now established – thanks to that technique. It could also tell us whether the paratacamite did, or didn’t have zinc in the lattice, which in turn narrows the likely region where the manuscript might have been incscribed.
In addition, the chemical profile set for colours used in a given culture, or environment is pretty definitive of origins for a painting, and McCrone is experienced in that side of things too.
I find it hard to envisage a failure to analyse the colours, as well as the inks, which could have told us so much about provenance.
Still, you could always just ask for a copy of the full report. What the Beinecke has published reads to me like part of the cover sheet /abstract to a lab report.
June 8th, 2011 at 10:42 am
I don’t know where else to put this – but linguists might like to look at the scripts on finds from the Hoq cave in Socotra.
June 8th, 2011 at 11:20 am
I’ve had another, slower read of the report. I think I’ve been unfair: the format is fair enough for a lay audience, and they had to allow, I suppose, for the amount of organics.
Did I mention that the analysis for the woman’s face is very interesting; copper with titanium etc.
June 8th, 2011 at 12:16 pm
Nick, Sorry – I’mprobably being a pest. But I thought you might like to see what such a report usually looks like.This one’s also relevant. The results from Beinecke 408 have other points in common with this map.
http://www.webexhibits.org/vinland/paper-clark02.html