Skies above. 6c: Methods, ideas and attitudes – and the ‘foreign’ in Voynich studies.

Header: (left) Precession and the northern circumpolar stars; (centre) night sky over Paris, Aug.22nd., 1420; (right) Precession and the southern circumpolar stars.

Two previous posts in this series:

update 11th November to show skies as calculated for Alexandria in 1420 AD, (August 22nd.).  readers comments and any correction most welcome.

The reason for spending so much time on the historical ‘backdrop’ is that when text and images are both problematic, as they are in the Voynich manuscript, we need to identify the ideas that will (one hopes) direct us towards the time, place and languages common to the first enuciator and his – presumably contemporary – audience.

How this is done: a thumnail guide to method and technique (1,500 wds)

As example: by merely looking at this small image (below, right) we might say that it is  “a bear, writing”.  That’s called treating an image as a “picture of..” Since the 1920s it has been the standard approach adopted by Voynich writers.

Against this approach, the analyst’s aims to – as it were – listen in on the communication  between the maker and his first audience, and assumes that their communication will be about their shared environment and languages – both verbal and visual.  We can do this for images produced in the pre-modern period because individual self-expression was not then perceived as the chief purpose of art  nor was ‘the artist’ the chief focus of attention.

So although this little detail could be imagined to be all about bears,  the form given it here depends on knowledge of [St.} Ursula‘s legend  (her name means a dear ‘little bear'(f.) combined with a specific error made by Geoffrey of Monmouth in relating her biography,  in twelfth century England. It is that error which is reflected in the conjunction of a little bear, and  writing.  (see details further below).

Recognising the tenor of this ‘conversation’, the researcher can now provide an upper (earlier) date for  first enunciation (a terminus a quo) and simultaneously identify a region within which maker and audience would be ‘speaking the same language’.

This, in turn, limits the range of spoken and written languages embodied in the accompanying text.  It must be one of those attested in the region and period where Geoffrey’s error had affect.

And this will be so  whether or not the accompanying text is legible.

Stylistics must then be taken into account. If they are not compatible with the information which an iconological analyst has ‘read’ from the image so far, that analyst must re-think the way they have read the image. It is quite unacceptable  to address opposition between the historical record and a personal impression by making statements which begin “the painter could have been/done…”.

In this case, the inclusion  of  a French-influenced bryony/’ivy’ border gives us the lower (i.e. later) limit for first enunciation: terminus ad quem.

The image cannot have been formed earlier than Geoffrey of Monmouth’s error;  the style of the bryony ornament makes it unlikely it was first enunciated later than the fourteenth century.

We can conclude with some confidence therefore –  whether or not the accompanying text is legible – that the manuscript’s content and specifically its written text came from a source available in the range 12thC-14thC and this detail was first given form in England and/or in 14thC France, a time when England and France were very closely connected by both popular and learned culture – and by politics.

I thus reach a conclusion that the ‘conversation’ between the first enunciator and his intended audience occurred in an Anglo-Norman environment and that it employs languages (visual and written-verbal) which were shared in that time and region:  Latin or one of the vernaculars employed at that time in England and/or in France would be indicated as the language of any accompanying text.

So now, having explained both  ‘why’ and ‘how’ the conclusions were reached, I must emphasise the the next step as the most vital when an opinion is to be shared, especially in a formal assessment of any object, and most particularly if (as is the case for most Voynich researchers) one has no access to raw data from laboratory tests or even to the physical object.

This step is where conclusions are tested against external scholarship and verifiable fact before being offered the reader, colleague or client.  And once again – if the evidence opposes an analyst’s opinion, it is they who must reconsider the matter.

To obscure disparities between one’s opinion and the objective historical record by creating stories or indulging in speculation is frowned upon,  and not least because it shifts attention from the object at issue,  to the researcher.   It alters a process of understanding to one of credulity.  It alters the relative roles of researcher and client or colleague – because instead of assisting their better understanding the object, it demands an act of faith from them to the researcher.  I should much prefer that a reader or colleague reacted to the information I provided by saying,  “I accept your evidence and understand your reasoning, but I won’t believe your conclusions” than “I don’t understand the thing any better, but I  will believe whatever you say.”

In this case, conclusions drawn from our reading of the ‘bear’ detail are very easily checked.  We have the manuscript’s catalogue record (Brit.Lib. MS  Egerton MS 3277), a very solid source because while no catalogue is perfect, it has always been the British Library’s practice  to exclude speculative matter.   Informed differences of opinion, where they exist, are always from well-informed persons; are clearly marked as items debated and the catalogue entry includes a bibiography which allows the reader to weigh the grounds on which each opinion was built.

So.. testing against the catalogue record (or other historical sources) shows that the ‘little bear’ on folio 13r is indeed in an Anglo-Norman work, one made in England,  and that the manuscript’s written text is in Latin and Anglo-Norman French, the whole manuscript being dated to the second half of 14th century.

Because the written text on folio 13r is legible, we can add further to our description and say the ‘little bear’ serves as illustration, specifically, for verses 6-8 of that Psalm. (Psalm 15 in the Vulgate; 16 in modern translations).

Here the iconological analyst would stop, even if privately they thought that, in addition to the rest, the image might also convey an indirect compliment to the scribe.

Why might the analyst think so?

In the western European Christian (Latins’) manuscript tradition, the usual order of production was that the scribe first ruled out the page and then inscribed the written part of the text. before the page was passed on down to the ‘pictors’ – whose available space, and its shapes, the scribe had effectively determined.  And  verse 6 of that Psalm reads:  “The boundary lines have fallen for me in pleasant places; surely I have a delightful inheritance..”  🙂

Why would the analyst not say this to the client or colleague?

Because there is no objective source against which that idea/insight can be tested. Simple as that. To prove an idea is to test it – to stress-test it – and in our discipline, the default must be that whatever cannot be tested must be presumed untrue. Others have different standards.


Monmouth and Ursula

In 12thC England, Geoffrey of Monmouth appears to have mis-read “Deo notus”(?) as Dianotus, and then supposed it the name of Ursula’s father. ‘Dia notus’ can be punningly rendered as a ‘record of days’ or ‘book of hours’ or even of the months (dian-notus).  On which point see entry ‘Saint Ursula’ in the Catholic Encyclopaedia of 1917 or the wiki article in which much of the same information is repeated. On ‘Dianotus’ see Geoffrey of Monmouth’s  (Historia.., Bk 5 Ch.14 ). Cusack credits a fifteenth-century Englishman, Edmund Hatfield with the form ‘Deonotus’, though Hatfield himself refers to a lost Latin sermon. In any case, Hatfield’s slip is clearly inspired by existing connection made between Ursula and literacy.

  • Carole Cusack, ‘Hagiography and History: the Legend of Saint Ursula’, in Carole M. Cusack and Peter Oldmeadow (eds), This Immense Panorama Studies in Honour of Eric J. Sharpe, [Sydney Studies in Religion 2], 1999, pp. 89-104.  (p.96)

There exists little on ‘the monk of Rochester’ Edmund Hatfield, though he is thought to have died in 1502.  My sources were

  • Cecil Henry Fielding (compiler), The Records of Rochester (1910)  p.246.
  • W.B. Rye, ‘Catalogue of the Library of the Priory of St.Andrew, Rochester AD.1202’, Archaeolgia Cantiana, Vol.3 (1860) pp. 47-64 (p.53). He notes the alternative spelling ‘Hatfeld’.

Though the explanation of Ursula’s true origin would take us too far from our subject, it is to the point that Ursula was associated with women’s education, and  in later Italy was chosen as role model (patron saint) for a community of nuns whose charter was to promote literacy and learning for women.

I hope I’ve shown in this example that analytical method is valuable whether or not the accompanying written text can be read.

Analysis of one small detail couldn’t provide the title of the text written on this page of Egerton 3277, not even the fact that it came from the Bible – though if one were to consider every image in that manuscript, or if one were a medieval person whose literacy began with memorising the Psalter, we might.  (Note that our aim is not only to say where and when a  manuscript, as object, was manufactured).

Today, what analysis of this type can do is to define the range in which those who are attempting to read an illegible text may concentrate their efforts with reasonable confidence.  The image of the little bear was fairly easily read but those in the  Voynich manuscript display a variety of influences and disparity between times and places of first enunciation. The work is clearly a compilation; its  images have been affected by time and by transmission. They incorporate evidence of distinctly different ‘conversations’  between one section and another, and even at times in a single page.  The whole then presents a fascinating range of questions which must be addressed first by analysis and then (as always) by ‘stress-testing’ any initial opinion or impression against the solid information provided by external scholarship.

Naturally, if one’s conclusions agree with those of an earlier Voynich writer, it is essential to credit that precedent – if you can be certain you have seen the original statement.


The language of art is compounded of a particular people’s shared culture, ideas and formalised conventions in expression – and of the verbal and visual languages proper to them.  These things together both inform and limit, first,  the mental image and then the range of its physical expression.

Because the imagery in the Voynich manuscript includes very few details exhibiting the customs of medieval western Europe’s common culture, we must work from a  wider historical ‘backdrop’ to identify the narrower historical context(s) which will make the manuscript’s content less unintelligible for a modern reader.

To this point, the backdrop now extends from the Hellenistic to the medieval centuries, and from Asia minor to the south-western Mediterranean and has shown how certain themes and concepts were maintained but variously expressed though that range.   What is now to be done is – so to speak – to move a problematic image across that backdrop until its form and content no longer appear remarkable.  It should look quite at home; still individual but not uncomfortably different, and most importantly no longer unreadable.

Connection to spoken language is a factor often overlooked in discussions of the imagery in this manuscript, though it is certainly true that when  the sense depends on some pun in the vernacular, or some event of only local and temporary interest, the meaning may be lost quite rapidly, even within the same traditions.

example – snails and knights.

Today many are puzzled by the frequent marginal images in books of hours where an armed knight is shown fighting a snail. Now, this might be an allusion to the centuries-long ongoing struggle called the  [re-]’conquista’ –  by way of allusion to Pliny’s term for a snail: concha.  But alternatively, or indeed, simultaneously, it may refer to the sort of things Crusaders had in mind when they set off to go a-conquering.  (to clarify this further might imperil the blog’s ‘G’ rating).

Or, again, it might be paralleling the adults’ battles with the children’s game of conkers, once played “using snail-shells, hazelnuts and the like (as Southey wrote in his memoirs in 1821).  Children’s games weren’t  of interest to medieval writers, and only Southey’s much later comment and the conjectured etymology offers support for that possibility.

[Conkers] have come from the dialect word for ‘hard nut’ (perhaps from the French for ‘a conch’ – ‘conque’), maybe from the old game using shells and nuts (‘conquerors’), or again from the French ‘cogner’ (to hit).

(I am indebted  to Jane Struthers’ blogpost (here) for that information.

Or, of course, it may allude (in addition or exclusively) to a pun on armour/armor – Amo(u)r vincit omnia.


The most problematic images – and the most intriguing for the specialist –  are ones whose  first enuciation clearly occurred in one period and culture, but which now include details indicative of  very different time(s) and attitudes. Images of this type bring  to the analyst’s calculations a  third, and dynamic, factor: affect from transmission.  Now, instead of a simple, linear structure for the research – such as that needed for the ‘bear’ in Egerton 3277 (see above ‘How this is done’) –  we have a sort of  historical 3-D chess problem with ‘transformations’ between one level (historical-cultural stratum) and the next.

This may occur when a new text conflicts with established conceptions of the world for a particular community, but we are principally concerned with the other side of that coin – when imagery itself was transmitted into evironments where it could not be ‘read’ as originally intended.  One must then take into account the probability that an original  detail  was accorded  different relative weight and value in one plane as against in the next.  Ideally, one aims to explain this too, but it isn’t always possible. History doesn’t always relate.

As example

we have a clearly Indian ‘Lakshmi’ statuette, of ivory, which was recovered from ruins of Roman Pompeii. How the native Romans interpreted it we don’t know: as the image of a slave, perhaps?  as the goddess ‘Venus? as an ordinary ‘dancing girl’? as the personal ‘idol’ of a bride brought from India? – or did some actually know the story of Lakshmi? History just doesn’t relate, and there is no basis for choosing one over another bit of guesswork..

Similarly, at present, we cannot explain the inclusion in the Rohan Hours of characteristically Buddhist-Hindu forms for the ring of guarding ‘angels’ on folio 159r, though the presence of similarly blue-faced angels in an Armenian church in Jerusalem suggests the idea and characteristics might, possibly, have arrived via Armenia. One cannot say – there’s no record, and too few examples in the western sphere to allow any sure conclusion. Interestingly, Armenian Christians had come, by this time (c.1420s) to bind their manuscripts in Latin style, i.e. with stitch-supports.

Problems of this kind are why researching really problematic imagery is the most fun for specialists of a certain type.  🙂

Some of the ‘angels’ still display the epicanthal fold, and most retain characteristically Buddhist rolls of fat on their necks. Some also display what we should call a double chin. One possibility among several is that the maker had seen some work in which an effort had been made to equate Christian with non-Christian ideas. Manichaeans and Nestorians did this and the first community to call itself Christian – the Community of Thomas in southern India – had believed it was founded in the 1stC by Christ’s apostle Thomas, a second wave of Christians arriving from Syria in about the third century owed allegiance to the ‘Nestorian’ patriarch, whose two capitals were in the fertile crescent.  All of which is suggestive, and nothing conclusive.

A further difficulty is presented when the receivers of transmitted imagery or text react negatively simply because information in it, or about it, seems to present an offence to their sense of what is personally right – their amour propre – and thus their allegiance to their own hierarchical ordering of persons or of ideas.

We have seen a hint of this mechanism at work in the way that Wilfrid Voynich, and even more William F. Friedman, approached the Voynich manuscript.

To either of them it was an idea intolerable (and thus instinctively seen as preposterous) that the manuscript’s content could be of non-European character and be a worthwhile study unless by, or at least mediated by, or owned by some high-ranking European male.

Just so, information and reasoning alone will not persuade a man who believes he has bought a seventeenth-century Cremona violin that he has an instrument sounding quite as fine but made by a nineteenth century emigre from Hungary.  His self-image is invested in the other idea, and since he is driven primarily by his beliefs, the only recourse is to refer him to some text, or person, in which he is predisposed to feel faith. He will often then accept ‘on faith’ precisely the same information.

History shows, repeatedly, that the strongest rejection of new information comes from the top and the bottom of the social hierarchy; from those with greatest self-regard and those who think truth is defined by “what everyone knows”.  Consider reactions to  the platypus, or to Darwin’s explanation of natural selection; or to a solution for the determination of Longitude.. This allows us to apply a (discardable) rule of thumb that ‘foreign matter’ will not have been first intended for the highest or lowest people in the new environment, something apparently borne out in the case of the Voynich manuscript by its materials, codicology and general presentation.

I do not think it true that personality-based decisions are necessarily a sign of the small-minded, but the pattern of history suggests that over-emphasis on personality has been one of the greatest hindrances to any intellectual advance.  It distorts the usual sense of the maxim that information is as good as its source.

Unhappily for the study, it seems that in William Friedman’s case, there was no person nor any academic field in which he placed more faith than he had in himself.  His aim was chiefly to prove his first ideas right and to ‘break’ the text.  I find no evidence  that he had any interest in the manuscript as such, nor troubled to learn anything much of manuscript studies, paleography, codicology, medieval art or even of medieval history.  And since he determined the line taken by his study groups, and thus the content, implicit biases and all, in d’Imperio’s Elegant Enigma, his approach deeply affected what followed after just as Wilfrid’s affected it throughout most of the twentieth century.

Happily, the study is moving on, but a milder version of that ‘Cremona vs Hungarian’ reaction is still apparent.

In illustration I’ll treat a couple of astronomical motifs from two folios in other parts of the manuscript.  The one offers a nice example of alteration – in this case addition – to its original diagram. The other appears scarcely affected by the process of transmission, and occurs in another ‘ladies’ section.  Both display knowledge of the southern skies – and that is the information which seems preposterous to some.

Stars beyond the book: Crux and false Crux in the Voynich manuscript.

In English writings, and specifically in Voynich studies, there lingers a habit of taking ‘Europe’ to mean the world, a habit still so general and so ingrained that one finds entirely nonsensical generalisations made, and regularly  assented to without pause for thought.

Scarcely an eyebrow is raised, for example, by such assertions as that  “the stars of Crux were lost” or   “most star-names are Arab star names” though even a moment’s pause for thought, or turning  to read even a wiki article, should have given that speaker pause.

For a bit of perspective, the introduction by Chamberlain and Young provides a pleasant first exposure to the wider view:

  • Von Del Chamberlain and M. Jane Young (eds), Songs from the Sky: Indigenous Astronomical and Cosmological Traditions of the World, pp. 49-64.(2005)

And for an idea of how complex even the idea of ‘Arab star-names’ is and what variety such imagery could take:

On the basis of accepting the bald statement that “the stars of Crux were lost” I have seen dismissed the fact that one or both of Crux and ‘false crux’ are depicted in the Voynich manuscript.

The unvoiced chain of thought among some theorists seems to run:, ‘Since my theory is inconsistent with depiction of Crux, therefore Crux cannot be depicted in the manuscript. I shall find some way to explain it which agrees with my theory, because my theory – which is mine – cannot be wrong.” (The usual alternative produced – or more exactly created as reaction to my first publishing this information –  has been Gregory of Tours‘ cross in Cygnus, but that has never marked ‘south’ in any system, not even the Europeans’.  Efforts to hastily manufacture ‘patches’ of this sort for a theory opposed by new evidence will normally forget to consider context, or to treat as one the critical elements of  context, function  and stylistics).

A more reasonable response in such a case might be:  ‘Since no reference to Crux has been noted in any European astronomical text dated before 1440, then if Crux is depicted in the Voynich manuscript, and if the manuscript was indeed manufactured in western Europe, this knowledge would have had to come from a different type of source –  written, oral or representational.”

To say that  ‘most star-names are Arab star-names’ is also – need one say – untrue.

The Arabs did have a name for Crux – Sulbar, meaning the beam of crucifixion, but this appears to be a result of early naturalisation of an older, Tamil term Sulba, meaning the knotted measuring cord.  The illustration shown (right) may serve as mnemonic.  It is from an old fresco showing a motif of older Nubian Christianity; cross and kombologion. Below (left) is Schiller’s (reversed) image of Sulbar, from one version of his astonishingly well-informed and constantly misinterpreted and underrated ‘Christianised Heavens’.

Sulbar – a binding and weight about the wrist of Abraham in one version of Schiller’s ‘Christianised’ Heavens’

Throughout that half of the world where it was seen each night, Crux served as practical marker of the night hours, and indicator of the unmarked Southern celestial Pole. In folio 67v-i it is used as an emblem for ‘South’ – and is complemented, correctly, by the other three astronomical emblems present in that diagram.  All four are actually superfluous – the directions already marked by other emblems – and employ a very different style of drawing to the rest.

More, while the assignments of its four asterisms are not wrong in astronomical terms they are a little odd in terms of traditional custom in those regions where Crux was a well known. The tradition of the eastern Arabs was to speak of Canopus as the proverbial ‘southern Pole’ for example.  So the four astronomical emblems on that diagram are a little odd – academic and  noticeably ‘foreign’ as if taken from a globe rather than from personal habit – but still, they are not wrong.  What they are is later additions.

I date their addition  to about the 13thC, not only – but not least  -because to represent stars as detached ‘heads’ is out of keeping with the regular practice of images in this manuscript.  Here I deal with only one of the four on folio 67v-i,  the emblem for south – Crux.(upper register in illustration, below).

False Crux (lower register in illustration above).

When I spoke first of the crosses in the manuscript, I identified the form on folio 79v-i again as Crux but going over these notes almost ten years later, I should now say this second is more likely meant for what we call  ‘false crux’ and which was certainly known, if confusedly, to medieval Europe.  The older pictorial traditions in depicting it may be divided into those which envisage it as the ‘shield of the face’ (I use the longer expression deliberately), and those which have it the cover of an entry to the world below and/or as a seal upon that entry from which the dead who are not truly dead will emerge again on the last day.

‘False crux’ lies in fact between Argo ratis and Canis major but is variously positioned in the European and Arabic imagery.

It has never been included among the official constellations but was very well known. In the older traditions it receives two interpretations, each given  various expressions.  In one strand, it appears as the cover of the cavern (‘mouth’) or entryway to the world below.  Bayer seems to have understood it so, though to have been uncertain about the difference between Crux and the false crux.

In the other strand of tradition, false crux is the seal and protection of the ship and may appear as a veil, shield other barrier set between the crew and the elements.  These images often betray uncertainty about how to show the ship, going ‘backwards’ can yet being drawn by Sirius (as Aratus say it is).

The asterism’s character –  if not its form – was clearly if surprisingly best understood in medieval Europe by the illustrator who created the following image, now in an early monastic copy of Cicero’s Aratea, Brit.Lib.Harley MS 647. That section’s manufacture is attributed to northern France, possibly Rheims, though to form images of words was not a custom native to Latin Europe, and is now associated earliest with works of roughly the same date made by the Karaite Jews of Syria (near Lake Tiberius) and of Egypt. And one mustn’t forget, since this is a copy of the Aratea, that when Harranian ‘star-worshippers’ were obliged to produce their holy books, or convert to Islam, or die during the first wave of Arab conquest, Aratus’ text was among those they produced, and their knowledge of astronomy was such that members of their community  established the study in Baghdad. In latin Europe this conception of Argo ratis soon devolves.

compare with

By contrast, a celestial globe made in Mashhad, as late as the seventeenth century, preserves memory of Argo as bird-headed, and more exactly here phoenix-headed and also of the ‘veil’ as shield against the dog whose rising theoretically marked Egypt’s annual inundation and drew the ship onwards (whether by stem or stern differs). This example takes additional sigificance from the fact that certain iconographic and stylistics found in the Voynich plant pictures occur also in a few leaves within the Mashhad Dioscorides.

In this connection, the comments made by Sadeh about the links between Mosul and Diyabakir, and use of parchment in the latter during the 12thC are of considerable interest, though the role of Nestorian and other Christian scholars in those regions, and specifically in connection with the transmission of knowledge about plants and medicines passes below Sadeh’s historical horizon.

  • M.M. Sadeh, The Arabic ‘Materia Medica’ of Dioscorides (1983) esp. Ch.2 (pp. 7-19)

About those stylistic connections to the plant-pictures in Beinecke MS 408, I’m can cite no prior ‘Voynichero’, and my sources were all academic ones. I daresday one might now find examples posterior to my study, and.or illustrations re-used illustrations from my posts to Voynichimagery


I repeat…


To help with orientation… The view from the northern hemisphere –  skies visible in Alexandria,  August 22nd., 1420 AD.

‘star atlas’ style…:

The glorious reality..

Postscript:  The next post, ‘Elevated souls Pt.1’ returns to the month folios. Once this series, focused on folio 70v-i ends, I;ll return to the short ‘reading-guide’ format with relief and pleasure which, I hope, my readers will share.

A much modified, ‘planispheric’, version of Schiller’s ‘Christianised Heavens’ can be seen here.



Skies above Pt 6a: Adding and removing layers

Header.  details from an Apulian terracotta, showing ten-rayed star; seven-rayed ‘star-flowers’, sun of night as fire-basket; clothed female figures.  (Metropolitan Museum of Art,. Dated third quarter of the 4thC BC;  inset – detail from an Attic krater, black figure ware.  Rape of Cassandra by Aias (Ajax) before the image of Athena.  (Met. Museum of Art, New York. Dated 6thC BC).

Two previous:


ALREADY, by this stage, a possible explanation had presented itself for the apparently deliberate avoidance of representing living things too realistically/literally.  The logs show that after noting a second exception (on f.116v)  I moved on to two more questions after putting notes in the margin to remind myself where to pick up later this matter of the  ‘boneless’ ladies** 

** vis: “Job 31:22.   Ezekiel 27:19; 29:6; 40:5-8; 41:8, 16-19.”

The two subsequent questions were 1. “Why female? why unclothed?” and 2. ‘why baskets? – March diagram. Significance?’. 

In fact I treated the second matter first, but will reverse the order here.  


“Why female? Why unclothed?”

Initial jottings:   ok. in Egypt.  Earlier Gk examples but in Mediterranean chiefly from c.2ndC BC. vide Pompeii; North Africa… Syria;  Black Sea, northern India and further). Western revival late – Renaissance latter half 15thC [cf. Panofsky and keeper of mss]. Female still  later.  Venice-Florence. .  

It was evident to me that the ladies’  first enunciation could not have occurred in an environment that was monotheist, aniconic or anti-iconic, nor within Latin Europe until a couple of generations later than the Voynich manuscript had been made.

It was equally clear that at some later stage they had been affected by such an environment. Logically, this indicated an impact during the period of transmission from when depiction of unclothed females might be of the generic sort (in the Mediterranean c. 2nd C BC – 5thC AD) to Europe where, early in the fifteenth-century, the present manuscript is thought to have been made but where depiction of unclothed female forms was still relatively rare and of ‘shapely’ ladies very rare indeed. Just as Panofsky observed in 1932. 

It not difficult to imagine a context in which an astronomical diagram of this type might have been first enunciated in medieval Latin Europe. But  I should think that imagination would be its only support.

So by lifting aside the intervening accretions (that is to say, the chronological strata), the figures in the month-diagrams can be seen again as shapely ladies with arms and shoulders whole, though without the still-later overlays of dress and heavy paint which a number of the deformed bodies now bear.

That the Voynich ‘ladies’ were originally unclothed, and their covering due to a relatively late sequence of additions is quite clear, the stages are exemplified below by details on folio 71v-ii.

In the first stage some additional lines were drawn,  details of breast and groin omitted or erased, and the body covered by a light wash. (below, left)

Heavier pigment was then applied, unevenly and not in all diagrams nor in every case.  The example (below, right) is again from f. 71v-ii. 

It seems to me that this heavier pigment which is consistently applied for the central emblems, was applied to the ladies as much in an effort to cover up their boneless limbs and ‘normalise’ the body as it was to render them more modest. 

 (The head wear is not peculiar to the western Mediterranean,  nor to the medieval centuries). 

But what this must imply is that the copy upon which those ‘improvers’ worked already showed the ladies with those boneless limbs – and so that the work had only recently entered their own horizons if the work was indeed done within western Europe.  In Latin Europe the usual practice was to eliminate or ‘correct’ images inconsistent with Latin customs or theology, and within a short time – an act of translation as the copies were made. 

In some few other folios we do see efforts at improvement/translation of the drawings per se, but it is noticeable that when such changes are very marked – as where a ruler was taken up to draw elements in the ‘bathy-‘ section – that hand does not remain present long.  The implication in that case, and where such details occur in other sections of this manuscript, is that the aim was to reproduce with near-facsimile exactitude, matter gained from one or more exemplars.  And then these drawings were ‘improved’ as if now under other guidance. The first wanted a precisely copied work; the other (whose effect is seen only in these month-diagrams) wanted something less awkward-looking.

Overall, I can only conclude that our present copy can be no original composition, nor any traditional product of medieval Latin culture.

Other items to be considered in connection with these figures (as I noted in the log) are that:

  1. the star-flowers may have seven points;
  2. a faint trace of ‘roses’ appears even in that altered figure from the ‘light April’ diagram. 
  3. ( certain of the baskets are drawn with a side-seam and in the detail from f. 71v shown above left, the basket might have two side-seams. A person doesn’t imagine such structural elements for an object if every object of that type they’ve ever seen lacks them.  So here the inclusion of a seam in the drawing is another technical detail, and another limiting factor, and all the more if these ‘baskets’ really came waist high: for ancient and medieval peoples say about 2’5″ (c. 62-63 cm).  [As it happened, that detail proved important]. 

A further question: Was the star itself identified only with the centre of these flower-like forms? To take the example of the aster, was it identified only with the disk flowers –  while the ray-flowers (‘petals’) were supposed… rays (Lat. radii)?  Implications for Q.20?


Why ladies?

The question  arises because, if the material had been transmitted through a community averse to realistic depiction of living things, surely they might have just omitted the ladies altogether, at very least from the month-diagrams. On the other hand, if their way had been informed by an idea that it was only a realistic or ‘workable’ likeness which was prohibited then the form itself and not the subject-matter was the focus of that prohibition.  In other words, their view was not that ‘you shall not make an  image in the likeness of any thing’ but ‘you shall not make an image as the likeness of any living thing’. So long as it did not actually imitate a living thing, you could make, or keep as many images as you liked.  This is also an attitude that speaks to an earlier time, when it was not uncommon to believe that a physical image and a living thing, both, had their animation or soul from the stars.  The soul was the star-soul, and it was drawn down into a perfected body – hence at birth.  Death saw that soul return to whence it came.  Something of this ancient belief remains today in the location imagined for a heaven of souls.  But there isn’t time to explain so much here, except to say that within the Mediterranean this idea was at least as old as the pyramids in Egypt.

That the figures weren’t omitted, but merely distorted might also, or alternatively, imply that the information was perceived as so important that only superficial changes were permitted.  Supposing them simply astronomical types in some sense, however, isn’t enough.  We have the model provided by some Hebrew- and Arabic-inscribed astrolabes to prove  that astronomical types could well be omitted without altering content.  But if each represented a personality (cf the Jerusalem Astrology, Sortes Sangalensis,  or the late ‘angels of the day’ in Agrippa) then they might be irreplaceable.  Or if each also was the token of a place and important for that reason when the original had been made.   But then why associate the stars predominantly with ‘ladies’?   One obvious possibility is that link of star-and place; with the tyche or other ‘patron’ – and some of the ladies in the month-diagrams and the bathy- section do wear battlemented crowns – technically ‘mural crowns’.  (This last was a matter which  I investigated a few months later than the work being discussed now. After I’d published some of my conclusions on that point, Koen Gheuns kindly cited my work when considering the same). Other Voynich writers have since referred to the same illustrations.

Below is a detail showing a star-holder, from an instrument probably made in Diyabakir, east of the Mediterranean coast. It was made during the lifetime of Roger Bacon and of an embassy which came to Europe representing both the Mongols and the Church of the East (‘the Nestorians’).  Diyabakir had been long been one of two major seats for the Nestorian patriarch (=pope), and remained so until 1402, when possession of the city and its surrounding lands was summarily handed over to the  Aq Qoyunlu  (the ‘white sheep’ Turcomans).   Eastern Greeks had  the astrolabe by c.200 BC



The ‘March’ diagram (f.70v) and the ‘classical nude’.

Reconsidering the ‘ladies’ on folio 70v as if with limbs restored to  “classical” form, the Mediterranean world provides only a  limited period for first enunciation of such figures and effectively excludes Greek art of the classical period – or rather that of indigenous origin.

Such a usage is commonplace in the Near East in the person of Astarte or Tanit, as well as in Egypt where it is quite normal for a fully frontal naked female to display her charms, but not in Greek art. In the 6th century [BC] it is only in extremis that women are portrayed naked or semi-naked, but this rule is broken in the case of hetairai, notably in two-dimensional art. Not until the end of the 5th century BC does mature female nudity begin to be emphasized.By contrast, it [was] acceptable in the minor arts imported from outside Greece to show the naked female form…

  • Alexandra Villing, ‘Naukratis: Greeks in Egypt’ paper published by the British Museum.

While the Greeks happily drew male figures without clothing, there are very few exceptions to the rule that the female body is clothed, whether it represents a divine or a mortal being. Four exceptions are (i) Aphrodite (ii) Cassandra (ii) bathing women (iv) heterai. We see no multiplicity of unclothed female forms depicted in these cases.

The Kneeling Bather.

Robert F. Sutton, Jr. has argued that in Greek art true ‘nudes’ occur from the  5thC BC with the type he calls the  ‘kneeling bather’.  However, as you may be able to see from the examples shown below, certain of his illustrations appear to show instead a  figure dressed in a short, wide-necked garment of about knee-length and even in these cases the loins are effectively covered.

  • Robert F. Sutton, Jr., The Nature and Function of Water, Baths, Bathing and Hygiene from Antiquity through the Renaissance.(2009). Includes a section entitled ‘Female Bathers And The Emergence Of The Female Nude In Greek Art’. 

Otherwise, the ‘untainted’ nude is usually dated from Praxiteles’ execution of his now-lost Aphrodite of Cos.  (I won’t include the controversial Ludovici throne).


The ‘tainted’ naked woman: Helen and Cassandra types.

Where Greek imagery does show a female naked, the implication is of a reduction in social status and/or moral standing.. or that the woman is bathing as preparation for marriage. The usual subjects are Helen (later of Troy) or Cassandra, who was raped by Aias (Ajax) in Athena’s sanctuary.  A woman was considered degraded by rape almost to the status of slave-prostitute. Volition had little bearing on that perception. And despite Augustine’s spirited theological argument to the contrary in his City of God (composed in the 5thC AD), the same ideas were characteristic of medieval Latin Europe.

In the first of the two examples shown below, you also see a characteristic that will later become rare, where figures are drawn with exaggerated buttocks but calves extremely slender: almost bone-thin.  Just a couple of images of Akhenaten and some statues recovered in Kiev ((10thC AD terminus ad quem) show comparable practice, but in the Egyptian and Kiev examples, the persons are clothed.

The Roman era saw Roman goddesses generally clothed, though a number of  Greek and foreign deities and ancillary figures (such as the Karites or Charites and followers of Dionysos) were divested of their garments, and we find various examples of the frankly pronographic within some frescos in Pompeii. Of the deities Aphrodite/Venus remains the exception to that general rule against full frontal nudiry.

The next image (below) dates to the 1stC AD, and is from Pompeii.  It shows a scene which by then had a long history in Mediterranean art, and with which many of the Roman military would be familiar.  After a ten-year siege, the Greeks entered Troy. At left, Menelaus reclaims his wife by raping her. The gesture of clutching a woman’s hair expresses both contempt and aggressive sexual desire and is maintained as a convention in art from the older Greek tradition.  On the right, the aged Priam watches helplessly as Aias (Ajax) rapes his daughter, Cassandra, within Athena’s temple.  The Romans believed their own capital city had been founded by the Trojans.  As you see, the form here given Cassandra’s body has much in common with the conventions of medieval Latin art  – a long slender torso and small, high-set breasts.

Erwin Panofsky rightly noted in 1932 that ‘shapely ladies’ do not appear in Latin art before c.1450. Even by that time,  the ‘renaissance’ movement involved only a small number of artists in a small region of Europe.  In that sense, the older historians of art were justified in considering the Renaissance a sixteenth-century phenomenon in art as distinct from a revial of interest in classical texts.

Michelangelo and Raphael are credited, for example, with having “initiated the practice of making preparatory studies of the nude prior to painting the figure fully clothed, in order to better understand the underlying structure of the body.”.   The opposite has occurred in the Voynich manuscript, where the clothing is added to obscure the drawing’s points of divergence from a perceived norm,  presumably the medieval Latin.   Michelangelo was not born until perhaps fifty years and more after the Voynich manuscript was made. He was born in 1475 and Raphael  in 1483. Both reached their maturity in the sixteenth century.

It is entirely usual – it has been usual for decades – that comparisons offered for imagery in the Voynich manuscript date to as much as half a century later than it was made.

As with a cat who brings only the mouse-tail, the evidence adduced to support most Eurocentric Voynich theories is  too slender, and too late.


Aphrodite: the nude bather type.

Even the Greeks’ Aphrodite was usually provided some covering until the time of Praxiteles’ remarkable work. But we cannot explain the ‘ladies’ in the March diagram as a multiplicity of Aphrodites. More to the point, there is no link between Aphrodite and the star in the Greek tradition, and in the Roman tradition only via their equation between Aphrodite and Venus.

But…to cut the longer story short for the benefit of my readers…  we do find Artemis  associated with a circle of lights, and with a sequence- and a circle- of moving women.

Just so, the stars form a series and a circle of lights.. and in this case  of women.

Artemis and her women.

a type for Artemis phosphoros

Their being depicted unclothed indicates, in my opinion, that we have either the usual Roman attitude to the gods of conquered peoples, or the different attitudes expressed by art of the Hellenistic east, and particularly  what emerged among Greek, Carian and Ionian on the eastern border of what had been the Persian empire, but which Alexander reached and settled.

I’ll return to that eastern sphere later.  In the next post I’ll look further at the ‘women of Artemis’.


[this might be good point to take a break, have a cup of something and remember to breathe.. 🙂 ]



Foreign deities, Christian Fathers, and Augustine’s  City of God (5th and 15thC AD).


Some years after the research was done that I’m tracking again now,  Ellie Velinksa wrote a blogpost ( Sept. 1st., 2013. ) in which she focused on a fifteenth century French version and translation of  Augustine’s City of God (The Hague, MMW, 10 A 11).  Taking a detail from its fol. 435 (above) she suggested that those  tiered figures offer a comparison for what we see in the Voynich month-diagrams, or more exactly those in which the figures are plain and unclothed.  The Hague ms was made half a century after the Vms’ radiocarbon dates of 1404-1438. The Hague ms is dated 1475-1480.What it shows is more that even so late as the last quarter of the fifteenth century, and even in a work produced by a master in France,  the medieval body is still the norm. with its long torso, waist not emphasised and  small, high-set, round (‘apple-shaped’) breasts. The detail selected by Ellie illustrates Augustine’s explaining the irrationality of astrology and the inconsistency of the different systems.It happened that in September 2013, while I was explaining for my own readers a certain technique (‘sieving’) used to find useful comparisons for works about which some certainty already existed, I cited the following image from a text composed c.1430.  It is fairly described as both English and French as the following details explain.

from ‘Medicine and Physiognomy from 14th to Early 16th Century’ (2004).  The passage quoted below is also available in French, translated by  Marilyn Nicoud and Nicolas Weill-Parot for Médiévales, No. 46 (2004/1) pp. 89-108.

Roland was of Portuguese ancestry through his mother. He appears for the first time as a student in the registers of the University of Paris for the academic year 1419-1420. After completing his studies, he became a medical master in 1424 and dean of the faculty in 1424-25 and 1427-1430 during the English occupation. His presence in Paris in the 1430s is evidenced by several indices, the most significant being his participation in 1436 in a dispute over the days of the year 1437 favourable to phlebotomy [blood-letting] and administration of laxative medicines. From 1436 to 1442 he was regent master, and died at an uncertain date in the 1470s (1470-1477) . Among the treaties attributed to him  is a manual of Aggregatorium sive compendium artis arismetice . The latter borrows long passages from Jean de Murs’ Quadripartitum numerorum and, to a lesser extent, Nicole Oresme’s Algorismus proporcionum , so it can hardly be considered an original work. He also wrote a work on geomancy which is novel in several respects, particularly the place that Roland grants astrology in that context. His Physiognomy was one of many works which were compiled or translated by the French members of the house of the Duke of Bedford, from already existing texts. Many of these translators or compilers were graduates of the University of Paris and they probably used the funds of the Louvre library. Roland dedicated the  Reductorium phisonomie to Duke John of Bedford in the early 1430. His work has survived in three manuscripts of the fifteenth century and a copy of the seventeenth century.

This reminds us that the nearest comparison from a European work for the Voynich plants’ style of drawing occurs in a medical manuscript made in England between 1375-1425 Brit.Lib. Sloane MS 335)  and that for the first half-century of its study, the Voynich manuscript was also believed to have been written in England by the Franciscan, Roger Bacon, who had also studied in Paris.


In the excitement of finding something similar to the images of Beinecke MS 408 in works produced in Latin Europe, it is very easy to overlook the many and obvious differences between the style of the one and of the other.  The Hague manuscript even more than Roland’s work employ a different palette, shows clear evidence of ruling out and – unlike the Voynich images – manifest the usual patterns of western Christian ideas about the world and its organisation.


On a personal note, Ellie has a good eye and one wonders what she might have discovered had she not come to the study with a ‘European-Christian-Duc de Berry’ theory, which limited the range of her work from the beginning.  She has surveyed only medieval Latin manuscripts in her work, as again in her hunt for ‘dotted stars’ (September 9th., 2013) and so remained unaware that the custom had an unbroken tradition through several thousand years, even in the Mediterranean.  Just by way of example, I’ve shown in the header a detail from ceiling of ‘dotted stars’ in a recently-uncovered tomb from pre-Ptolemaic Egypt.


Augustine and polytheism in the 5thC AD

While Augustine lived, the Mediterranean’s polytheistic heritage was still a living culture, one with which he was entirely familiar having converted to Christianity only in his late maturity.

Augustine was a north African, Phoenician by descent. After having been interested in Manichaeism, which was then a widespread religion across the southern Mediterranean including Byzantine north Africa, Augustine became a Christian as his mother had long wished he would.  A thousand years before the Voynich manuscript, he composed a work entitled The City of God. In fifteenth-century Europe, its message gained renewed point; those complaints voiced by Romans of the 5thC AD about the loss of Rome to barbarians were being paralleled by the Byzantines now seeking refuge in the Latin west.  In each case, there was a suggestion that the city had been lost because the older ways in religion had been abandoned.

That argument, and Augustine’s reply, may have led to the work’s renewed popularity among the non-clerical class, but certainly a magnificent copy and translation into French was made in late fifteenth century. The painter is believed the same Maitre Francois who is credited with that image of Orata whose discussion began the present series of posts.

His City of God begins with reproaches to those Romans who, by pretending to be Christians and taingn refuge in Christian churches, had been spared by the Goths during the sack of Rome. Unlike Ajax, these ‘barbarians’ respected the right of sanctuary.  Augustine knows well the classical gods and classical poets.  Nor was he alone among those whose works preserved that knowledge within Latin Europe.

Tatian, an Assyrian Christian who wrote in Syriac, had addressed the Greeks in the 2ndC AD, saying:

Be not, O Greeks, so very hostilely disposed towards the Barbarians, nor look with ill will on their opinions. For which of your institutions has not been derived from the Barbarians? The most eminent of the Telmessians invented the art of divining by dreams; the Carians, that of prognosticating by the stars; the Phrygians and the most ancient Isaurians, augury by the flight of birds; the Cyprians, the art of inspecting victims. To the Babylonians you owe astronomy; to the Persians, magic; to the Egyptians, geometry; to the Phœnicians, instruction by alphabetic writing. Cease, then, to miscall these imitations inventions of your own. Orpheus, again, taught you poetry and song; from him, too, you learned the mysteries. The Tuscans taught you the plastic art; from the annals of the Egyptians you learned to write history;

  In the same century, Clement of Alexandria could discourse on the subject of human sacrifice, mentioning Tauric Artemis as he did:

The Taurians, the people who inhabit the Tauric Chersonese, sacrifice to the Tauric Artemis immediately whatever strangers they lay hands on on their coasts who have been east adrift on the sea. …  Monimus relates, in his treatise on marvels, that at Pella, in Thessaly, a man of Achaia was slain in sacrifice to Peleus and Chiron. That the Lyctii, who are a Cretan race, slew men in sacrifice to Zeus, Anticlides shows in his ‘Homeward Journeys’; and that the Lesbians offered the like sacrifice to Dionysus, is said by Dosidas. The Phocæans also (for I will not pass over such as they are), Pythocles informs us in his third book, ‘On Concord’, offer a man as a burnt-sacrifice to the Taurian Artemis.

The cult of Artemis in Tauris had been known to the Greeks as early as the 5thC BC, when Euripides told the story of Iphigenia in his plays.  Having been saved from becoming a sacrifice at the hands of her father, Menelaus, ‘Iphigenia was taken by Artemis to serve as priestess in the goddess’ temple among the Taureans, where as part of her office she was obliged to sacrifice strangers thrown up upon that shore.  The latter part of her story is in Euripides‘ ‘Iphigenia among the Taureans’ ( Ἰφιγένεια ἐν Ταύροις)’.

Chersonese Taurica: floor mosaic (3rd-2nd century BC) Encyc. of Ukraine
Chersonese Taurica: floor mosaic (3rd-2nd century BC) Encyc. of Ukraine

Contemporary with that mosaic is the small terracotta figure – findplace unstated – shown below.


In the mid-thirteenth century, still, a temple of some sort stood  on an island off the coast. In c.1245, it was mentioned by William of Rubruck (Willem van Ruysbroeck), who said:

So we made sail for the province of Gazaria, or Cassaria, which is about triangular in shape, having on its west side a city called Kersona [ancient Chersonesos or Cherson; mod. Sevestopol] … And as we were sailing past Kersona we saw an island on which is a temple said to have been built by angelic hands.

The Tauric Chersonnese is now a UNESCO heritage site).

This might be a good place for images of those figures recovered from Kiev and certainly made before the mid-tenth century AD.

Nor was Augustine, in fifth-century North Africa unaware of how an unclothed woman looked.  Apart from his own experience – he had a long-term companion and several children – but from mosaics and other forms of imagery, of which a remarkable amount has survived until today from the time of Roman occupation.  Here again, however, the proportions of the body are much closer to those of medieval European art than to the Voynich manuscript’s ‘ladies’.

detail of a mosaic. Roman period. North Africa. Courtesy of
naked, nude, un-clothed.  

I don’t describe the anthropoform figures in the month-diagrams as naked, or as nude.  The words carry overtones that I do not think can be applied yet to those figures.

Naked carries an implication of force, and chiefly of male force. A sword is naked, not nude; a body is found ‘naked’ not nude.  Unless the subject is an infant.

‘Nude’ implies a voluntary  disrobing, and a subject physically embodied.   Cassandra is naked; Helen is usually and Aphrodite invariably nude when depicted without clothing.  English doesn’t have a  neutral word for the situation we have here, where clothes may be irrelevant – so ‘unclothed’ will have to do.

The Romans’ interest in ‘shapely’ female forms ended as monotheism rose to dominate the Mediterranean.   A partial exception is found in some astronomical images, and in certain medical works which would emerge, in Latin Europe, from the schools of Paris during the 1400s.  But here again – as almost always – they offer no close comparison to what we find in the Voynich manuscript.  First enunciation of its ‘ladies’ occurred, in my opinion, during the Hellenistic period though derived ‘organically’ from earlier roots – and not necessarily Greek ones.


“They have no speech, they use no words; no sound is heard from them. Yet their voice goes out into all the earth, their words to the ends of the world”.