Appraisals – pictures

I have found no comment or appraisal earlier than the 1930s which suggests that the manuscript’s vellum, inks or writing style was thought to offer any contradiction to Wilfrid’s firm conviction that it had an English provenance.

About the imagery, all commentators before Erwin Panofsky seem to have  agreed it looked thirteenth-century or so, or simply shaken their heads in bemusement. It was recognised as manifestly not within the range of  ‘normal’ western Christian medieval art.

And Panofsky would say the same, shortly after Wilfrid’s death.

The now constantly re-iterated assertion of that the manuscript has a “central European” origin is an idea that arose quite late and owes less to the primary document, or to Panofsky’s answers to Friedman’s questionnaire, than to exactly the same combination of gut-feeling and dubious hearsay, combined with hero-worship, which always characterised Wilfrid’s own position. These were ideas imposed on, not extracted from, the primary document. In each case, proponents of ‘theories’ developed a priori have ignored evidence or opinion pointing to a contrary conclusion: that’s how parallel universes begin.

It has been suggested pretty recently that the style of earlier [re-]binding supports the ‘central European’ idea, but a re-binding in about the seventeenth century might be anticipated, because Barsch mentions specifically having a few scribes copy several sections for Athanasius Kircher’s advice.  Bindings were not then treated with the reverence they are today, and a binding’s date or place doesn’t always help provenance content or origins.

Back again to the first decades of the twentieth century.  Wilfrid still has the floor.  As far as he believes, or any of his associates will say, the manuscript is thirteenth-century English; an original authorial creation, its written text enciphered  by that one Christian male author imagined to be of ‘scientific’ bent.

Nobody tried to fit the manuscript’s imagery into that scenario, but then imagery in this case was not being treated in the usual way: as a primary indicator of provenance. Mr. Pelling’s comment (see below) has made me realise that I  expressed myself badly in that sentence. In fact the problem was that Newbold (and perhaps others of whom I’m not aware) *did* try to fit the imagery into their scenarios, but did not test those scenarios against any separate investigation of the imagery, and in the usual way the imagery in any text of unknown provenance, especially one in an unknown script or language, serves as a primary indicator of the entire work’s likely provenance.  (there – I hope that’s clearer.)

A lot of the usual processes were not adopted, perhaps because so few of those willing to become involved with ‘the Voynich manuscript’ had much competence in that area.  The most extraordinary thing is that when it became clear that the manuscript didn’t find any parallels in the world of Roger Bacon, nobody looked much further. No-one began by comparing its imagery with precedents outside the Latin manuscript tradition, and certainly not beyond the Mediterranean.

There is no explanation for this failure: it simply wasn’t done.

Later, when opinions were sought from relevant experts, the trend was towards the southern part of Europe, probably Spain.  But again, it was only Panofsky who suggested a non-Latin environment and his opinion was “tactfully ignored”. It went against the desire of the majority (mostly specialists in English literature who had connections to the military and intelligence sectors) that the work should be ‘scientific’, Christian, and in some sense or other “quasi-Protestant”. That’s how it was back then.

The few, apart from Panofsky (who appears after Wilfrid’s death) but who were genuinely expert in any area relevant to the manuscript’s imagery, didn’t much like what they saw. It just didn’t fit the western, Latin, tradition.  Nor the Greek, nor the ‘Arab’, actually.

In illustration, I’ll refer to the now-famous revulsion which was expressed by T.A. Sprague, a genuinely qualified and experienced specialist in western manuscript traditions for illustrating Dioscorides’ Materia Medica. When approached – decades later – by Brigadier John Tiltman and shown photostats of some plants from the manuscript, Sprague was rendered speechless, repelled and apparently left at a complete loss to explain them.  Tiltman records the incident:

In 1957 I paid visits to a few specialists in early herbals in England. Among them I saw the late Dr. T. A. Sprague in Cheltenham and showed him a few specimen photostats of herbal drawings from the Voynich manuscript, of which he had been previously unaware. As he looked at them he became more and more agitated and eventually said,  “Do you know what you are asking me to do’? I have spent the last twenty years of my life trying to identify the plant drawings in the Juliana Anicia codex when the names of the plants are given in Greek, Latin and usually Arabic, and you are asking me to identify these awful pictures.”

“awful” pictures – not ‘amateurish’ nor ‘inept’ nor ‘childish’ nor ‘insane’ nor ‘tenth-century’ nor ‘Leechbook’ nor “German’ nor ‘Spanish’…  Just “awful”.

The response was at once a genuine reaction, and face-saving excuse for it, by an expert unable to get his footing when confronted by the Voynich imagery. Sprague had no reason to study, or recognise, styles of botanical imagery other than the considerable corpus of those Greek,  Latin and Arabic manuscripts.


What is also noticeable is that, despite their floundering, none of the older experts suggests  the manuscript was fake.

That last idea is now the gut-feeling (or ‘theory’) most championed by Richard Santacoloma, providing a weaker but genuine counterweight to the twin themes of  “central European” theory and imperial ownership: championed chiefly by Rene Zandbergen and the focus of all his work in Voynich studies.

As with Wilfrid’s dedication to the  ‘Roger Bacon cipher’ scenario, where he constantly introduced the manuscript as the ‘Roger Bacon Cypher Manuscript’, there is a regular tendency to hope that an ability to convince enough people – to ‘have the numbers’ may see one preferred scenario drive any others to the margin.  That process does not normally require much by way of evidence gained from a primary document, though presenting unremarkable data in plenty, and the odd conclusion in the right tone of voice, can have strongly persuasive effect.

The Voynich fake

Not that there weren’t fakes about in the early twentieth century.  One certainly fooled Wilfrid Voynich, who bought it, and the British Library when they bought it from him in 1905.  The item is still in the British Library, catalogued as B.L. Add MS 37177.

The following paragraphs come from an article written in 1968 for the British Museum Quarterly, where B.L. Add MS 37177 is among the works attributed to a nineteenth-century forger described as “the Spanish Forger”.

In what follows, I have moved one sentence forward (italicised).

The identity of the ‘Spanish Forger’ remains obstinately shrouded in mystery, although his existence was recognized as long ago as 1930. [An extensive list of published examples was compiled by Otto Kurz for inclusion in his book on fakes in 1948] and some of [the ‘Spanish forger’s] work is demonstrably at least seventy years old. … He owes his name to Belle da Costa Greene of the Pierpont Morgan Library, who began to collect references to his work some forty years ago. Her original list, now kept up by John Plummer, has been increased to a total of about forty-six items, of which fifteen are panel paintings, twenty-five are separate miniatures on loose leaves of vellum, and the remaining six are manuscripts containing more than one miniature… . Several further examples have been brought to the Museum for examination during the past four or five years, and two fine specimens were sold at Sotheby’s in July last year….

… another [example of that forger’s work] was purchased as genuine as long ago as 1905 (Add. MS. 37177) [it is a] miniature at the British Museum catalogued as representing Hernando Cortes landing in Mexico in 1519…Fortunately two letters concerning the purchase of this particular item are preserved in the departmental correspondence for 1905. It came from a London dealer, W. M. Voynich of Shaftesbury Avenue, who offered it for £75 (though he would have asked 1oo guineas from a member of the public) and said that he had bought it from another bookseller, at a high price, as a picture of the landing of Columbus.

Here’s the picture:

Hernando_Cortes_landing_in_Mexico,_16th_C_-_BL_Add_MS_37177 forgery

Backhouse continues:

He [Voynich] himself suggested that it might in fact be meant for Cortes, probably on account of the rather eccentric coat-of-arms on the flag, which appears to quarter the arms of Castile, Aragon, Granada, and Navarre. Navarre was not in Spanish hands until 1512 and Columbus’s voyage took place in 1492. Voynich thought that the miniature might have been a frontispiece to Cortis’s official report to the King of Spain, but he had been unable to trace such a manuscript in any Spanish archive. It passed the scrutiny of G. F. Warner, but he did question its provenance and Voynich wrote again to say that it had come to England from a dealer in the south of France. He seemed to have had it either from a Basque or from a ‘Polish Count’, but the English bookseller who sold it to Voynich had said that the transaction was so long ago that the French dealer’s name and address were no longer available. This provenance, particularly the reference to a Polish count, is not very convincing.

A source for this miniature has since then been identified. The figures are grouped in the same way as in plate ix of T. de Bry’s America, part iv, first published in 1594, and the ships [in the miniature] appear to be modelled on plate xii in the same book. It is worth noting that the forger has altered the costumes, which were contemporary with the date of the book’s publication, to a style more in keeping with the date of the actual event. The de Bry woodcuts illustrate the landing of Columbus and it is clear from a sketch map on the back of the miniature, also spurious,* that Columbus is in fact portrayed, in spite of the flag. The map shows the Atlantic Ocean and the West Indies, and the Fortunate Isles (Canaries), San Salvador (Watling Island), Cuba, and Haiti, all visited by Columbus on his historic voyage, are particularly marked…

* an aspect of the map not mentioned in the British Library catalogue entry, and no image of that map appears to have been made available to the public as yet. -D.N. O’D.

Paintings by the ‘Spanish Forger’ are usually dated to the late nineteenth or early twentieth century. [emphasis by present author]. The British Museum’s Columbus miniature seems to have made the earliest documented appearance, in 1905, even if the time which is supposed to have elapsed between its purchase in France and its sale to Voynich is not taken into consideration… Certainly [the ‘Spanish forger’] had every intention of deceiving would-be purchasers…His attempts at making his paintings look old are, at least to an amateur’s eye, fairly convincing… Most convincing to a layman is the obvious antiquity of most of the vellum. Some pieces, including [another attributed to the same hand] Dr. Millar’s, seem to be blank leaves removed from manuscripts and sometimes have ruling on the verso. More frequently the miniature is painted on a leaf or cutting from a large late medieval choir book, the music having been removed from one side to make a space for it. It is particularly interesting to note that Add. MS. 53783 and the two historiated initials in Philadelphia are painted on pieces of the same manuscript, as well as sharing similar border decoration.

Janet Backhouse, ‘The “Spanish Forger”‘, The British Museum Quarterly, Vol. 33, No. 1/2 (Autumn, 1968), pp. 65-71

But the punchline to this story is not that Rich Santacoloma, talking to the strangely closed ‘parallel universe’ of Voynich studies, has had to argue for years that old parchment might be used by a forger. The wider world is well aware of such practice. No, the point here is that – notwithstanding the many people who have classified the miniature of B.L. Add MS 37177 as work of “The Spanish Forger” – the British Library catalogue remains unchanged (apparently) since its original inscription in 1905.

More than this, there was an exhibition held at the British Museum for which a catalogue (1990) had been prepared, evidently to the very last proofs, and which had included that ‘Columbus/Cortes’ miniature in the list of works by the nineteenth-century “Spanish Forger”.

But when the catalogue was printed in that year, by the Trustees of the British Museum and the University of California, clear marks remained of what must have been a very last-minute editing of the text. And the subject of that editing was the miniature sold to Library by Voynich: B.L. Add MS 37177.

The picture itself is not included, yet its original reference and caption-line remain.  It was to have been the third illustration of the “Spanish Forger”s works, and bore the illustration number 201(c). You can read the whole book,  Fake? The Art of Deception through the Internet archive.

The following clips have come from p.189 and p.191. No pages are missing. What came to be missing was any reference to B.L. Add MS 37177 as being a nineteenth century forgery by the “Spanish forger”. The Spanish forger’s works are (or were) grouped under the same number 201.  The ‘Columbus/Cortes’ image was to be the third example 201(c).  What happened? Did some uber-expert change everyone’s mind? Not that I can discover.  There just seems to be a quick erasure of the last line from the captions, removal of the image, and hasty addition to the text suggesting that the ‘New World’ image was by some unspecified and undated hand.

Forgery clipsp.189

Spanish forger clip2



For me, though perhaps I’m too idealistic, the really staggering fact in all this is that there is no mention whatever in the British Library’s online catalogue that the work is a forgery, by an earlier or by a later hand. it looks as if the entry remains exactly as it was first written in 1905.  What is more, there is no mention of the map on the reverse as fraudulent (which all agree it is). I should have liked to see that map, but I’ve never been able to find an image of it.

Now, you see. It isn’t only Voynich studies where this sort of thing happens: evidence, no matter how expert, can always be ignored.

Here’s the B.L. catalogue entry as of today:

Add MS 37177 : 16th century

Title: A MINIATURE in colours apparently representing the landing of Hernando Cortés in Mexico in 1519; with a foliated border on two sides, containing a second portrait of Cortés to the shoulders, within a medallion. The actual miniature measures 6 1/8 in. by 4 1/4 in., and it was probably, prefixed to some nearly contemporary account of the Spanish conquest of Mexico.(!!) The style is Spanish, and the artist must either have personally seen natives of the country and its forms of vegetation or have had before him some picture painted on the spot. On the back of the leaf is a small map of the west coast of Europe and Africa and the east coast of Central America, with the names in Latin. Vellum leaf, 9 3/4 in. x 6 in. XVI. cent.

Creation Date: 16th century
Extent: 1 item
Conditions of Use: Letter of introduction required to view this manuscript.
Contents:  Hernando Cortés, conqueror of Mexico: Miniature of his landing in Mexico: in 1519.

Art. Illuminations and Drawings SPANISH: Miniature of the landing of Cortés in Mexico: in 1519.

 (Those measurements in millimeters: 155.57 mm x 107.95 mm).

Puzzled about why this should happen, or why the 1990 catalogue might be edited at the last moment?  I am too. And not that I want to make more of co-incidence than it deserves, and getting a story right can certainly take a bit of time, but as it happens, American audiences have always hunted for ways to present the manuscript in a more ‘transatlantic’ manner;  it must have pleased them to read in 2004 a screaming, if minor, headline in England’s Daily Mail which read:

Has the Voynich manuscript been decoded? Mysterious 15th century text may be written in a lost AZTEC language

  • The Voynich manuscript was discovered in an Italian monastery in 1912
  • Due to its location, historians think the manuscript was written in Europe.
  • It is full of illustrations, diagrams and a mysterious text written left to right.
  • Cryptographers have been trying to decipher this text for decades
  • Botanist now [sic.] claims the plants in the book come from Mexico.
  • This suggests the book may be written in an Aztec language called Nahuatl.

This may explain one very peculiar aspect of the tests for which a German television company paid McCrone to perform some years ago.  They had the tests run against a sample of the rare (and highly unlikely)  mopa-mopa resin, whose only source in the world is a very small region in south America. Not surprisingly, McCrone found no mopa-mopa in our manuscript.

So it looks as if that pot has been some while a-boiling.  And if you’re a television story-maker, it would certainly be an enormous pity not to be able to make use of such a perfectly iconic image as a picture of some Spaniard (maybe Columbus, maybe Cortes..) standing among a lot of undressed women, one you could link directly to Wilfrid Voynich.  Not the same impact if it’s by a nineteenth century forger –  wouldn’t you agree?

And though pictures of nekkid leddies undoubtedly grab the viewers, the style of painting here proves… nothing in common at all with the content or forms in the Voynich manuscript.


Stern Christian seaman being adored by lots of foreign women

Stern and righteous Spanish rock-star being greeted by hordes of melting fans …. right…


  1. Diane: I was always under the impression that there were quite a few people who didn’t buy in to Wilfrid Voynich’s hopeful history prior to 1930. John Matthews Manly for one is a definite: Charles Singer too?


    • Do you mind if I leave that question for the moment? It’s one thing for people to have an opinion earlier than they express it – another simply to have doubts which become firmer with time. So I hope to deal with the evolution of those peoples’ ideas as I map the progress of “Voynich studies”. The next post (two if I must) will outline public statements made about the ms before Newbold’s death, and later to 1931.
      At present I’m still in the immediate post-war period… with some digressions.


  2. Ah – your comment is about my first sentence, I think? I’m speaking of the manuscript as such there: “I have found no comment or appraisal earlier than the 1930s which suggests that the manuscript’s vellum, inks or writing style…”etc. I am excepting the marginalia and the added inscriptions on the month roundels. The first time I’ve found them mentioned – including by Manly – is a little later, but I do have to rely largely on what I can access online.


    • Diane: I have some copies of correspondence between Voynich, Newbold and Manly from the 1920s that I photographed in the UPenn archives back in 2006, that makes what was going on back then a bit clearer. I’ll try to tidy them up and post them on Cipher Mysteries.


      • I look forward to it. I have also been writing around, trying to see if anyone holding Panofsky’s archives (they’re split) can find notes to which one of the obituaries refers. They say that he left “some thoughts about” the manuscript.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s