Starting from Scratch – #8 correlating expert and scientific assessments

Voynich coarse vellum

“The vellum is coarse, even for the thirteenth century” – Robert Steele (1928) [1]

[This is very long. Please feel free to print it off and read at your leisure.  Or, if invited, I’ll deliver the paper at the next Voynich conference. rofl]

Materials vs content.

Fascinating though the subject of the Voynich script is, and for all that I should include for comparison some from beyond the  Mediterranean, I’ve decided to move on with  codicological and iconographic evidence.

This is, after all, a hypothetical exercise, envisioning the direction in which this study might have moved from the first, if Wilfrid Voynich had not been so effective in promoting his romantic – and  largely imaginary – narrative for the manuscript’s Latin Christian origin, authorship and subsequent history.  Despite his failure to provide evidence for it, Wilfrid’s story was adopted in his lifetime and assumed true for the following seventy years. To this day, most researchers presume that most of its details (if not his posited author) are proven.

I have said that I will follow this trail, setting aside more recent opinions including those gained from my own research, but I admit this is proving difficult when reference to a previous post here  is the easiest form of elucidation.

Without investigating scripts used beyond the Mediterranean we have at least noted a potential discrepancy between (i) the time and place the artefact was made and (ii) those for composition of matter contained –  which indicates at the very least an earlier recension.  As for manufacture, our default remains, for want of better knowledge, the Veneto or thereabouts, early in the fifteenth century.  The script’s movement and posture being unlike Latin scripts of that region and time, we find it most like the earlier Spanish Sephardic hand and/or early fifteenth century Greek cursive. In either case, one has to doubt the usual assumption of an entirely western and Latin content.

Not only by interpretation of the material evidence have we reached this point, but by reference to the early evaluations of the manuscript, these having been relatively independent and less dictated by the assumptions inherent in Wilfrid Voynich’s romantic provenance.

Certain customs in the Voynich imagery – including deformation of anthropoform figures  – add to the indications of non-Latin, and non-fifteenth century roots for the manuscript’s informing text or texts, at least in those sections where such figures occur.

The manuscript’s fold-outs are another unusual feature, sitting awkwardly in the context of Latin bookmaking to that time – so they will be the next point of comparative study (see two posts following).

Considering the present popularity of a late-come theory of a German Latin origins, it is surprising that we have seen nothing so far which might  be said to support it. Neither the work’s dimensions, nor its page layout, its hand, or anything else has spoken to a German provenance – so far. As we proceed, the situation may change but to date our evidence suggests strongly that we owe the manuscript’s content to nowhere further north than Spain or the Balearics (though southern Byzantine regions,[2] such as Crete and/or Egypt remain possible).  The artefact’s formation we can attribute, so far, to the early fifteenth century, and northern Italy.

Baresch and Panofsky

All of which conforms with Georg Baresch’s belief that the content was not Latin but ‘oriental’ – and he had the manuscript in his possession for forty years. Again, it again both confirms, and explains, Erwin Panofsky’s first assessment given to Anne Nills in 1931, reported by her immediately in a letter to her friend Herbert Garland. In part:

A certain Dr. Erwin Panofsky .. is at present in New York and Miss Greene suggested that she bring him and Mrs. Voynich together – very decent of her don’t you think. So Mrs. Voynich met him at the Morgan Library where she showed him the photostats (note that they are negatives and now in poor condition, having greatly faded in some parts). He became intensely interested and seemed to think the MS. early, perhaps as early as the 13th century. He asked to see the original, which we showed to him last Friday. His first impression was that it was early, but as he came to the female figures (in conjunction with the colors used in the manuscript) he came to the conclusion that it could not be earlier than the 15th century! .. Furthermore he is convinced that the MS. is Spanish (or something southern near Spain) and shows strong Arabic and Jewish influences.[3]

As readers of my blog will know, we have since then gained knowledge of a mid-fourteenth century Jewish work written in an early Spanish hand, in which ‘shapely ladies’ of closely comparable form appear, and which derives from a tenth-century work, but more importantly from a version said specifically not to descend from the Latin corpus.[4]

Robert Steele

Our manuscript’s palette (or some substances within it) now remains as our only evidence that any additions might have been made to the content in the fifteenth century – something which again agrees with early evaluations of the manuscript’s appearance. As example, Robert Steele although he supposed it a Latin work said he thought it ‘strange’ that, if it were deemed later than the thirteenth century,

“the draughtsman should have so completely escaped all medieval or Renaissance [Latin] influences.”[5]

So now – if we posit an effort by some fifteenth-century copyists to reproduce earlier matter, and an effort made on their part to keep as close as possible to the form of their exemplars (leaving aside the addition of colour), then it would explain the absence of influence from the Renaissance, and why the great majority of early appraisals found little reason to object to the thirteenth-century dating which Wilfrid Voynich argued (reasonably enough), but which then became the basis for his creative story about the manuscript’s supposed authorship and history – a story which in the absence of evidence must be regarded as largely fictional.

In other words, no one leapt up in 1921 and said, “pshaw! thirteenth century English, nothing. This is a fifteenth century German (or Italian) Renaissance work” – did they?  And not because they were ignorant of fifteenth century Renaissance works, but simply because –  as Panofsky, Steele and other professionals  saw immediately –  the work presents as one which, if originating in Europe, would be from the twelfth or thirteenth century – save for those sections containing ‘shapely ladies’ and only Panofsky drew attention to that issue.

So this becomes an interesting point: given this evidence that much about MS Beinecke 408 presents as not later than thirteenth-century, with only those sections containing ‘shapely ladies’ supposed (at present) from the fourteenth, so nothing remains to require our supposing that the written part of the text composed any later than the mid-fourteenth century, and much to suggest that it may be considerably older.

Radiocarbon dating three folios from the first six quires returned dates from the fifteenth century, so we may tentatively take that as indicating the time of manufacture, this being (apart from any marginalia or later re-binding) the putative final stage in the artefact’s evolution. It is a curious fact that the usual stratification of a manuscript – manufacture as distinct from first composition of the content – has rarely been part of discussion in Voynich studies, though in manuscript studies the distinction is routine.

So deeply did Wilfrid’s notion that the work was an original, authorial, composition sink into the consciousness of his own generation that few have even paused to consider since then that a manuscript made in the fifteenth century is very likely indeed to contain content first composed elsewhere, and rather earlier.

I expect these implications about the contained matter might be difficult for some cryptanalysts to accept, for it brings the possibility that our written text which has so far defeated all efforts to understand it,  might have gained its form not in the fifteenth but as early as the thirteenth or fourteenth centuries if not earlier, a time when the art of cipher among Latins was fairly rudimentary and from which no system of formal encryption is known which would so constantly defeat modern cryptanalysis as the Voynich text continues to do. To quote Philip Neal’s comment on Rugg’s proposition of a pseudo-text,

It has been argued – I used to argue myself – that the [Voynich] phonetic structure was beyond the powers of a 16th-century forger to create, so that the text must be a real language or an unknown type of cipher.

Mary d’Imperio, Friedman and Tiltman,

That systems of true encryption existed earlier, elsewhere, we know.  On the other hand, Friedman and Tiltman who seem never to have spent much time establishing whether or not the work originated in Europe’s Latin society, appear to have presumed the point already established by Wilfrid i.e. they adopted the view that the work was the original composition of a Latin author.  Each then concluded –  within those parameters – that the text is not enciphered but  closer in type to an invented ‘synthetic language’ and/or ‘universal language’, medieval examples of which are referenced in  d’Imperio’s Elegant Enigma.

d’Imperio’s book does not refer anywhere to codicology, though Section 11.1  contains a well-phrased plea for more careful attention to be paid the manuscript as material object:

In my opinion, it is of primary importance that the inks, pigments, and vellum of the manuscript be tested and examined scientifically and compared to those of other manuscripts by paleographers [sic] and art historians… As far as I have been able to discover, no such research has ever been carried out. Further, there are no current plans on the part of the present owner of the manuscript (the Beinecke Library at Yale) to make any such studies in the near future. Nevertheless, only studies such as these can offer any hope of satisfactory answers to many of our questions. They could turn up crucial new information that might completely alter the complexion of the problem. I hope that some present or future students will be able to arouse interest in a scientific physical study of the manuscript.. .  [6]

Even within a hypothetical framework, approaching the manuscript in that way does indeed completely alter the complexion of the problem.  What we have gained from the manuscript’s materials and imagery so far is that the work is not an authorial text, but a compilation; that it was not first composed in the fifteenth century, or in northern or central Europe, but towards the south, possibly in Spain, by not later than the mid-fourteenth century –  even including its ‘shapely ladies’.  This accords with other internal evidence from the manuscript, co-incides with the time of the papacy’s residence in Avignon (1304-1377) and  with a new flourishing in Genoa and in the Balearics of maritime cartography – a genre to which I have concluded that some of the manuscript shows connection.

The end of that period is also marked by the requisition by the king of Aragon of a substantial library of manuscripts which had been acquired by a Jewish physician born in  the mountains above Dalmatia, a scholar and traveller known to the Latins as “Leo the Greek”.  In the year of his death, which was the year of the papacy’s return to Rome, there was an unusual amount of book- chart- and document-copying being done, one which illustrates the range and nature of the parchminers’ network of that time.  I have already written on the last topic in 2010 – 11, in my old research blog and again here.[7]


I do not think there is sufficient reason any longer to argue against the work as a compilation from earlier sources but the next question is then –  to what purpose?



1.Robert Steele, ‘Science in Medieval Cipher’, Nature 122, (13 October 1928)  pp. 563-565 .

2. on Byzantine bindings, see Szirmai, The Archaeology of Medieval Bookbinding [available online as pdf] or its chapter on Byzantine bindings reprinted here. For Byzantine and Greco-Latin bindings in Europe, Silvia Pugliese’s paper ‘Byzantine Bindings in the Marciana National Library’ originally published in The Book in Byzantium: Byzantine and Post-Byzantine Book-Binding can now be downloaded through

3. Passage from Anne Nill’s letter to Garland is quoted from the transcription which was made from the originals and published by Richard Santacoloma in  ‘Anne Nill Speaks’ (blogpost), 16th January 2013.

4. speaking of the manuscript MS Sasson [Sassoon] 823, now  UPenn MS LJS57, the relevant series of drawings “obviously follow the style of those which accompanied the catalogue of al-Sufi [but for reasons explained in their article] it would appear that the drawings cannot have been copied from the Sufi latinus corpus”  adding later that although the hand is ‘early’ Spanish [~Jewish] – “… one ought to consider the possibility of Spanish influence on manuscripts executed at Prague…  the first known astronomer in Prague was sent as a ‘gift’ by Alfonzo of Castille to Premsyl Ottokar II King of Bohemia”.   Karl A. F. Fischer, Paul Kunitzsch and Y. Tzvi Langermann, “The Hebrew Astronomical Codex MS. Sassoon 823″ The Jewish Quarterly Review , New Series, Vol. 78, No. 3/4 (Jan. – Apr., 1988), pp. 253-292.

As noted in my previous posts where that manuscript’s imagery is mentioned, Langermann describe that manuscript as made as a collection of extracts from older works; on vellum; with quires of 8 pages each. The dimensions are given by the current holder, the Lawrence J. Schoenerg Collection as 275 x 204 (190 x 132) mm. bound to 288 x 224 mm. Full catalogue details online. Note that the membrane is there described as parchment.

5. Robert Steele, op.cit.

6. Mary d’Imperio, An Elegant Enigma.  Invented and universal languages, see  § 9.3 “Pasigraphy”. Her appeal for closer study of the manuscript as material object, §11.1 “Paleographic and Other Scientific Studies of the Manuscript”.

7. My posts and comments refer chiefly to a paper by Dom Anselm M. Albareda, “The Preservation and Reproduction of the Manuscripts of the Vatican Library through the Centuries”, Proceedings of the American Philosophical Society, Vol. 104, No. 4, Dedication of Library Hall of the American Philosophical Society, Autumn General Meeting, November, 1959 (Aug. 15, 1960), pp. 413-418. [JSTOR]

Note – in 1988, Jim Reeds confirmed an observation made earlier by [?] that the British Library contains  “Facs 461 [which] is a collection of old photostats of folios 1-56 from MS Beinecke 408, as well as Facs 439 which contains more photostats, various articles about the manuscript from the period 1921-22, and a number of letters addressed to Robert Steele in which the manuscript is discussed. These include letters written to Steele in 1921 from Wilfrid Voynich and from William R. Newbold.  Jim noted especially two letters from A. W. Pollard to Steele, written on British Museum letterhead, and which “deplore Newbold’s work” according to Jim.

A pencilled draft of a paper by Steele was also among those papers.

One would dearly love to know more.


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