Swallowtails – foreword.

Some time ago I quoted SirHubert’s comment:

.. breaking ciphers is all about testing hypotheses and finding *the* consistent solution, of which there will be only one. Historical research doesn’t admit of one neat solution and works very differently.

“SirHubert” comment to ciphermysteries, (December 10, 2013)

To unpack it a little, we can speak of ‘left-brain’ and ‘right brain’ strengths. The following comes from a discussion about computer programs, but it’s nicely short and clear.

The left brain is the bully brain. It doesn’t just complicate things with its logic, it goes one step further. It drowns out the free-thinking nature of the right brain. But first let’s deal with logic.

The left brain is mathematical and logical. So like all maths problems it likes to be correct every time. Which is fine when you’re dealing with maths and 7 + 3=10 (and can never be 11). Every thing has to be black and white.

It’s different when you’re drawing, or playing music or writing an article. You can have your black and whites and a range of rainbow colours. This of course drives your bully brain totally crazy. It’s trying desperately to pigeon-hole what you do into black and white. And of course, it fails. (Sean d’Souza)

Sometimes the right brain is described as the ‘creative’ or as the ‘feminine’ and the left brain as the ‘logical’ or ‘masculine’.

We all have both a left- and a right-brain, and ideally they should complement one another, but for historical reasons modern western society associated ‘left brain’ behaviour with ‘the superior sort’ – as a cultural, gender, national typecasting – and right-brain behaviour with ‘inferior types’ – as individuals or as groups. You know the sort of thing – ‘Foreigners/women are so very emotional’… ‘their sect of Christianity is superstitious; ours is rational’.

Throughout Mary d’Imperio’s Elegant Enigma you can see an (often-unconscious) suppression of the ‘inferior’ in the attitudes expressed both to people and to ideas, to the point where even wildest flights of unsupported guesswork and ‘historical’ fantasy are presented as if they were a left-brain product: that is, supposedly ‘logical’ inferences. I’ve provided examples in earlier posts about the Friedmans and d’Imperio’s book.

This confused mentality, which mis-identifies ideas springing from the imagination and presents them as if a product of logic, are the worst type to have engage with and attempt to discuss images.

True left-brainers have problems. They often develop enormous ‘blind spots’ in their idea of reality. They look for, and like, ‘the similar’ while tending to dislike and even not recognise ‘the different’. Right-brain skills are the most helpful when it comes to understanding images of any sort, but for rigorous analysis, both left- and right- should operate in concert.

Much of Wilfrid Voynich’s seminal talk of 1921 was of the ‘confused’ type – imagination presented as if the product of solid information and well-informed logic. He spins chains of speculation and sets historical fact side-by-side with his fictional history for the manuscript. The implied congruencies are an illusion.

The problem of misrepresenting imagination as ‘logic’ thus began early to infect studies of Beinecke MS 408. It was not only that people presented a baseless fiction as if a product of logic, but that they, themselves, believed it was so.

Because they believed their logical left-brain skills had produced an idea, and not their inventive right-brain skills, requests for the evidence and for an explanation of their reasoning were often met by silence, by dismissive noises, contempt and/or expressions of indignation.

If you think that’s an odd reaction for ‘left-brainers’ – because mathematics, computer programming and other ‘left-brain’ occupations expect any problem to be presented with both its solution and its working-out, I can only suggest that image-making and historical studies are regarded as easy, as ‘soft science’ and ‘mere right-brain stuff’ by the sort of person who identifies as a pure left-brainer.

So, while it is possible for some mathematicians to just gaze at the sky and produce the answer to a long and difficult set of equations, if you ask them to provide the evidence and a logical exposition, the best will do so without raising an eyebrow. None of the ‘how-dare-you-question-me’ about such experts. Of course if the person asking is a six-year old who has yet to hear the word algebra, the expert may decide to condense. The point is that an iconological analyst will offer the same depth of explanation too.

Provision of an answer with its working-out is also how the written part of the Voynich text is usually discussed – statistics and arguments from those statistics informing any posited ‘answer’. But the infuriating thing is that the very same people often accept (and what is worse, produce) very sloppy work when the subject shifts to the manuscript’s drawings.

It is as if they had an ‘either/or’ attitude, and switched off their analytical left-brain the moment the subject changed. Most show no sign of understanding the range of preliminary studies, or the appropriate methods needed to read accurately pictures made at least six hundred years ago, and whose time and place of first enunciation are as yet unknown.

Robert S. Brumbaugh is a good example of a person who regarded himself as a superior type -a ‘left-brainer’. If you read his essays, you might try to identify the ideas he adopts without evidence adduced; the extent to which he relies on imagination; on others’ untested assertions; the absence of any reference to histories of medieval and earlier art; the absence of analytical and comparative evidence in his exposition of one drawing or another. My own impression is that he had no idea how to determine whether a particular detail in the manuscript was intended to be read literally, metaphorically, symbolically, or allegorically.

His ‘blind spots’ are often the same as those informing the Friedmans’ work, and d’Imperio’s account of it.

Extreme left-brainers are comfortable with quantification, and least comfortable with matters requiring balanced judgement of qualities. They can easily compare things in pictures – “these are pictures of cats” – but are at a loss when asked, for example, to decide and explain which of the pictures best conveys ‘feline nature’ – and how the maker did that.

In my experience, as I say, it is not the natural ‘left-brainers’, nor predominantly ‘right-brain’ types who are worst at reading pre-modern pictures. It is those ‘confused’ ones who identify as left-brainers and logicians but whose logic is badly flawed and who constantly mis-interpret, and so mis-represent the products of their imagination as products of evidence-dependent logic.

Often perceptive in social situations, they make very poor analysts. In fact, if they manage to qualify, they can prove a real menace to the profession because they deceive their clients.

Being able to convince themselves that anything produced by their mind must be a product of ‘logic’, but also adept at persuasion, they persuade their clients to believe that some bit of their own historical-fantasy is a valid description of the object for whose assessment the client is paying. The hardest sentence for such types to utter is: ‘I can offer no informed comment’.

Otherwise, most people can learn to develop the skills natural to both left- and right- brain. The right-brainer is the most direct and acute in observation and more interested in learning more, the left-brainer more inclined to shy away from the uncomfortable or hitherto unfamiliar.

That may seem counter-intuitive but imagine the situation – it’s the middle of the night. The person awakes with an impression that they’ve heard a noise. The right-brain says ‘burglars’ and wants to investigate to see whether or not that is so. The left-brain says ‘Nonsense – probably the dog playing with its ball’ and wants to go back to sleep. A balanced mind might – might – say ‘It might be burglars; it might just be the dog’, then curb any impulse to go downstairs but instead pick up their phone, move into a room with a lock on the door, and notify the police. That sort of reaction is not wholly impulsive, nor wholly ‘logical’ but it is eminently reasonable.

Most people not in the grip of some theory can be taught to recognise the difference between a painting made in fifteenth-century Venice and a manuscript illumination made in thirteenth-century Constantinople. They can memorise the tokens which distinguish one saint from another in Christian art. But the ‘confused’ types won’t want to do the necessary reading; they’ll want quite complex matter reduced to bullet-point slides – everything ‘right’ and simple. I’m sure you’ve met the type. Not stupid, exactly, but not clever in the right way.

The sort of question which seems to bring their mind to a complete stop are ones requiring informed, but qualitative judgement, such as: “what inference would you take from the use of pink in a fifteenth century painting?”, or if you simply ask them to ‘read and explain’ a particular picture, they cannot do it. The fear of ‘being wrong’ – as ‘not logical’ – is overwhelming because they suppose the definition of the ‘superior’ left-brainer is ‘never wrong’. Oddly enough I’ve found that many can relate well to images of the most overtly literal propagandist sort, and are most comfortable with early twentieth century poster-art of that type.

In my experience, when the questions mentioned above are asked, such persons habitually either produce an answer at random and when asked to explain it say something like, ‘It stands to reason’ or turn towards some other person for an indication of the ‘right’ answer.

Extreme left-brainers have persistent difficulties if the aim is to understand without producing any sort of ‘answer’. It’s good training, though, because it helps break the habit of assuming any question is a ‘problem’ in need of solution.

One encounters in Voynich studies, but less often in daily work, the curious situation where a left-brainer simply presumes that the ‘right answer’ can be determined by a simple head-count. This is much what Friedman did in issuing his ‘questionnaire’ or when Voynicheros use meaningless phrases such as ‘generally accepted’ or ‘not generally accepted’ to obscure an inability to offer an informed opinion of their own. The obvious, if impolite, rejoinder is ‘Sez who?’ and the scholarly one, ‘Why?’

For a century it was ‘generally accepted’ that the work was an autograph written by Roger Bacon. In 2011, promotion of ‘seventeenth-century Prague’ theory had seen the manuscript’s dating ‘generally accepted’ as sixteenth or seventeenth century and its subject ‘generally accepted’ as about pharmacy and alchemy. None of it was a result of consensus among formally qualified specialists, and dissenters’ views were considered outliers. Until the radiocarbon dating confirmed the informed consensus reported decades earlier by Kraus’ agent.

Truth-by-numbers may seem ‘democratic’ but it’s rubbish when the numbers are a random collection of persons, persons repeating untested ideas, and the subject is a medieval manuscript.

One of the questions which, over the years, has produced many interesting and illuminating answers as (non-Voynich) students were being encouraged develop both left- and right- brain skills, may also interest some present readers as something to mull over:

“Would you describe the doctrine of papal infallibility as the product of left-brain or of right-brain thinking?”

Hint: if your first instinct is to turn towards some other person and try to get ‘the answer’ from them, consider yourself one of the ‘confused’. If your first instinct is to learn as much as you can about the history of arguments about papal infallibility before saying anything, your right-brain skills are probably well developed. If your first instinct is to ask, ‘What has papal infallibility got to do with  pictures?’ then you’re probably a left-brainer with a lot of reading ahead of you.


Voynich ‘ciphertext’ attracts left-brainers.

When Wilfrid Voynich found the manuscript and realised he could read neither the written nor the pictorial text, he arbitrarily defined the written text as a ‘ciphertext’ just as he arbitrarily assigned the whole work to a single author, imagining ‘logically’ the imagined author of this imagined ciphertext must be Roger Bacon, and when he arrived in America, he tried first to promote the ‘ciphertext’ as potentially valuable for the military.

The result was that Wilfrid became a person of interest, as possible subversive, and a file was opened by the BOI, precursor to the FBI. That response, by the way, is a nice example of how logic may operate without evidence, and unchecked by reason.

So Wilfrid then changed tack and began emphasising the ‘Roger Bacon, scientific genius’ line, as you see from his talk in 1921.

But in a sense, it was too late.

Cryptography and military matters in general thereafter provided, with computer programmers, most of those interested in the manuscript, and such professions tend to attract ‘left-brainers’. While I’d certainly not deny that cryptographers and mathematicians are capable of imaginative leaps, they tend to be uncomfortable with situations where a question simply doesn’t have any yes/no ‘answer’ and the value of an opinion is the range and quality of an individual’s understanding.

This is why, I think, Friedman imagined that the Voynich text could be approached as contemporary ciphers were. He presumed a consistent orthography, an ‘official’ grammar, and that the aim of any enciphered text was to send a disguised but neat, monolingual, ‘plain text’ to persons at a distance from the person who composed a message.

None of those assumptions – including the assumption of encipherment – had been a conclusion from solid historical evidence, but it explains why Friedman saw nothing ridiculous about presenting Erwin Panofsky with a pre-determined set of seventeen ‘questions’, the aim of which was to collate all such answers as if the number of ‘yeas’ and ‘nays’ could decide whether an opinion was or wasn’t right.

I think it also explains why Friedman was unable to give Panofsky’s first opinion (given to the Voynichs via Anne Nill) the weight it deserved. He didn’t know how to decide relative weight due one ‘expert’ opinion over another.

For more than a century, the idea of the Voynich text as a ‘ciphertext’ became one of the study’s ‘doctrines’ and as far as I can determine no-one formally qualified to analyse images had contributed to the study at all between 1932 and 2009. It was a ‘left-brain’ field.

By 2009, it had become another “Voynich meme” that “until the written text is deciphered, any comment about the drawings is subjective”.

Like so many other Voynich memes, it’s rubbish, of course, but a perfect example of ‘left-brain’ irrationality.

First, as a moment’s thought will show, every major city in the world has art galleries and museums filled with items that were not produced and appreciated by means of any explanatory screed, and secondly because the implicit, and unrecognised ‘given’ behind that dictum is that we may impose the habits of modern western art on these pre-modern drawings.

It’s not unreasonable to approach a piece of modern art with the expectation that it might be described by one of just two categories namely, ‘portraits of things’, or “subjective expression of an artist’s worldview”. In the first case, the artist is imagined ‘drawing what s/he sees’ and this is supposed to be accessible to every viewer. In the second case, the viewer supposes that they can only access meaning by resorting to their own emotional responses and/or reading a written description.

But just as Newbold was wrong to suppose the medieval pharmacy like an early-twentieth century American drugstore, so this notion of image-making isn’t appropriate to the pre-modern world or to non-European tradition.

As well as ‘portraits’ of things (pictures ‘of’), and individual ‘expressions of ideas’ there were ‘pictures about’ and they were ‘about’ the things known and shared by the first maker of an image and his intended audience. In a different time and a very different environment.

Which is why the habit of ‘matching pictures’ by defining all pictures in terms of a single object in them is so prevalent a habit in modern Voynich writings, and so very inappropriate without the provision of historical and technical explanations of precisely how and why the images juxtaposed should be considered ‘alike’.

Here’s a concrete example of why simplistic ‘picture matching’ is useless to explain any image included in such composites.

Quite apart from a need to recognise and describe stylistics – which tell the informed viewer that the pictures (from left to right) are Japanese, Egyptian, and German – and without the background studies of history which will allow the analyst to offer an informed opinion that the first is probably seventeenth-century AD, the second probably second millennium BC and the third, fifteenth-century AD, the iconological analyst must be able to key those matters to the most vital issue – intended significance.

That sort of work needs both left-brain and right-brain skills, in coordination.

Readers should be alert to alleged ‘comparisons’ relying on the viewer to invent a ‘logical’ link to justify side-by-side juxtapositions and be cautious about accepting tacit arguments from such asserted ‘similarity’.

It is not unreasonable to ask a theorist presenting such efforts at silent persuasion: ‘Why do you think so?’ ‘Where’s the evidence?’ ‘What are your precedents’? ‘Explain your thinking’ – such questions are ones that professionals expect any client or any fellow to ask, and which is addressed, as a matter of routine, in any written report.

What happens in Voynich studies, all too often, would be unacceptable in any other context.

If a person who inherited a picture comes asking for its explanation, you don’t say ‘its a man on a horse’, lay it beside a picture of a mounted Napoleon, and tacitly invite the client to infer that their picture should be dated to the nineteenth century and supposed French. But the equivalent is constantly done in Voynich studies and, by long usage, that appalling habit has come to be regarded as normal methodology in this study.

In the next post I’ll reproduce some of the earliest contributions to discussion of the ‘swallowtail merlons’. I’d like readers to consider what ‘givens’ are assumed, what ‘blind spots’ they can identify, what basic errors of reasoning are evident, and whether they think the topic deserves a thorough re-consideration.

What magic? Where magic – 4.2 Whose magic? Byzantium – Spain.


old Toledo

Questa citá di Tolleto solea
tenere studio di negromanzia;
quivi di magica arte si leggea
publicamente e di quiromanzia;
e molti geomanti sempre avea,
esperimenti assai d’idromanzia;
e d’altre false openion di sciochini,
comm’e fatture o spesso batter gliorehi.

Luigi Pulci (1432-1484?)

Only this town of Toledo/holds classes in necromancy;/there you can read about magical arts publicly – and chiromancy;/ and there numerous geomancers demonstrate experiments in hydromancy;/ and other false and foolish notions ..



Constantinople medieval reconstructed birds sml

When emperors accuse courtiers of making them sick through demonic magic, and  make use of astrology when making important decisions … when manuscripts of sorcery that require extremely high levels of erudition are copied and employed, and when senior churchmen are accused for using, and actually being, practitioners of magic, it is quite clear that what is being dealt with here is not to be dismissed as “superstition” as the misguided, ignorant and unrepresentative beliefs of a lowly social group or a few isolated individuals, but something that was an integral part of general Byzantine culture and thought.” (pp.151-2)

  • Richard P.H. Greenfield, ‘A Contribution to the Study of Palaeologan Magic’, in Henry McGuire (ed.), Byzantine Magic, Dumbarton Oaks (1995)  pp.117-153.


To anyone thinking of crafting a ‘magical’ theory for the much imposed-upon Voynich manuscript, my advice is  – Don’t do it.

What follows in the next few paragraphs is editorial comment. I’d usually make it optional, collapsed text, but till wordpress’s new block editor provides that function, readers who can’t be bothered with editorials can just scroll down to Part 2.

Reasons for advising against creating a ‘magical’ theory include:
1.  ‘Does the manuscript consent?’  Seriously.  Fictional-theoretical narratives have been imposed on this manuscript, one after another, for a century with many based on no more than subjective impressions of one or two drawings – drawings which they do not understand, and do not attempt to learn how to analyse.  It is better to work from the manuscript’s evidence, and to first investigate whether some specific characteristic of script or codicology points to that subject, or whether there is any clear parallel between the Voynich drawings and any known traditions in that type of image-making.  The great error in the history of this study has ever been that a person moves from researching the object to researching the fascinating and comfortable ideas produced by their own imagination. They cease to be researchers and become, in effect, novelists.
2. ‘Magical’ texts and images are a highly specialised area of scholarship. To get some idea of what a mass of preliminary study you’d need to have under your belt if you hope to say anything useful, try reading Peter Forshaw’s thesis. He has  posted it in separate chapters at academia.edu.  You might pay attention to his curriculum vitae while you’re there. His overview is in – Chapter 2.
    • Peter J. Forshaw, ‘The Occult Middle Ages.’
3. Cyber-bullying.  In thirteen years of observing the behaviour of online ‘Voynicheros’ I have found only one theory-group which actively tries to deter researchers by ad.hominem pack-attacks, and that is the ‘Germanic-central-European-Christian-nobility theory’ group.
The true focus of their interest is not so much the manuscript as an idealised image of the Rudolfine court as being, somehow, the quintessence of ‘Germanic suavity’.  Apart from Toresella, who has a faintly ‘down and dirty’ idea of magic, the only material which that group will countenance must be consonant with keeping lace collars and cuffs nice and clean, and preferably Protestant or at least quasi-protestant. So the not-terribly-occult theme of Astrology, and the really-quite-gentlemanly Alchemy are the two forms of occult learning they tend to impose on the manuscript, although – as a simple matter of fact – the Voynich diagrams do not conform to the traditions of central European image-making in either subject.  I agree that there is some circumstantial evidence for thinking that Georg Baresch believed that the manuscript in his possession probably related in some way to some form of alchemy.
However the great flaw in that ‘Germanic-central-European-Christian-gentlemen’ theory is that the primary document withholds its consent.  There is nothing about the manuscript save a bit of marginalia which speaks in any way to a German impact on the text.  The codicology offers no particular support for a Germanic theory; nor does the binding, nor the page layout, nor the style of script (notably lacking the strong vertical emphasis of German scripts), nor the drawings, though by considering none but German-and-central-European manuscripts, ignoring the existence of any but supportive opinions, and by using a very lax system of ‘compared images’ an impression of validity has been presented, fairly successfully, to an uncritical audience.
If you develop a theory about, say, Spanish magic, or Aegean magic, or non-Christian magic for the manuscript, members of that now-dominant group will either ignore you, or attempt to get you to stop your own line of research and devote yourself to that theory, or – worst case – will harass you by constantly ‘meme-making’ as a means to impugn your motives, intelligence, qualifications and even your mental and moral soundness.  You cannot have a reasonable debate with adherents to that theory; their theory is never presented as a formal thesis,  and the view held in common is that to engage in factual debate, or to engage with any dissenter is beneath their dignity – though continual avoidance of the objective issues and relentless ad.hominem attacks is apparently ok, because a dissenter is – in their view – a ‘lesser person’ by definition.  Not everyone who adheres to that theory is so unreasonable, but enough are to ensure that members toe the line.  The arrogance of that group has grown to a point where one core-member recently repeated another smart-sounding meme to the effect that any non-believer, regardless of their position in the world of non-Voynich scholarship, is a ‘maverick’ for declining to serve that theory.  Which just shows just how badly divorced from reality any mutually-reinforcing team can become.
What I find sad is that a number of that group are individually intelligent, reasonable and highly competent in some relevant discipline.  One can only wonder what the study has lost by their conformity to a theory untenable by any normal standards.
Belonging has definite advantages – so long as you limit your work to the perimeter defined by ‘ western Christian nobleman of Germany-and-central-Europe’, all will be warmth, good fellowship and shared sniggers at the ‘others’ in any surviving Voynich arena online.  Your work will receive many appreciative comments, regardless of how ridiculous your ideas might seem to an outsider, someone like – just for example – Peter Forshaw.
You may also like to consider the ethics involved now that some members of that theory-group have  moved beyond merely refusing to acknowledge the existence of informed dissent, and have begun actively erasing mention of such persons and research from supposedly objective histories of the study on websites and Voynich wiki articles.  A recent example of this heightened folly occurred recently in regard to a scholar named Rainer Hannig.
It is not the point that his ideas were incorrect – or even correct.  The point is that the history of this study since the rise of the Prinke-Zandbergen theory about twenty years ago has been ‘fixed’ by tweaking or even inventing information, and by dividing all acknowledged information into two groups – the ‘sensible people’ who support that theory – and all the rest.
If the aim were to erase all matter not a validated and solid contribution to the study, error-free, then we should have to erase everything except the scientific analyses and Prescott Currier’s talk of 1967 1976.
And that’s why I don’t encourage you to create a ‘magic theory’: the manuscript does not invite it; there is a strong likelihood that you won’t have time to learn enough to say anything of lasting value unless you already have years of specialised study behind you –  and even so, if your research and conclusions oppose the Prinke-Zandbergen storyline, you and your research are likely to be ignored and/or attacked ad.hominem and/or retrospectively ‘eliminated’ from the study’s history.

So now, having been clear about the inadvisability of following that line –  let’s move forward.


Part 2.

We pick up from where the last post left off.

In that, I offered some few items in evidence for 14-page quires (septenions) having been used in fifteenth-century manuscripts from Byzantium, Italy and Spain, and further that ten-page quires (quinions) which are not quite so uncommon, also survive certainly from fifteenth-century Italy.

The Voynich manuscript’s Quire 20 was originally a septenion; of its quires Quire 8 and 13, one originally was, and the other still is a quinion.

If we consider Lascaris’ book-collecting journeys in search of classical texts, together with the distribution pattern for Hebrew manuscripts which Beit Arié records for  septenions, it becomes clear that their incidence relates to the maritime routes which connected the north-east to the far south-west of the greater Mediterranean. (I’ve added a star for north Africa, not for its septenions, but as a centre of magical practices which influenced both Spain and Sicily.)

mediterranean-map transmission points

This in turn tells us that the routes are those over which Venice and Genoa held control for much of the medieval period, and until the fifteenth century.  .

It is evident that Lascaris travelled the Venetians’ route when he went to acquire copies of ancient and classical texts, but the Jewish examples, cited by Beit Arié for the western Mediterranean, lie on the routes controlled by Genoa.  This is understandable since in Genoa itself, as in the Genoese ‘colonies’ in Constantinople (and Pera), as in Caffa on the Black Sea, Jews and Genoese regularly worked together.  (If anyone wants references for this, other than what can be found online, email me.)

Northern Italy, lying between those routes, was open to influence arriving from either side and in fact Italy’s Adriatic coast was where many foreign enclaves were established, including eastern Christians arriving from the eastern Mediterranean coast.

Venetian and Genoese trade routes medieval and Trebizond

To recap Janus Lascaris’ journey, in the third quarter of the fifteenth century: he took ship from Venice (Padua being in the Veneto), for..  Corfu, Arta, Salonica, Galata, Sozopolis, Athos, Salonica, Crete, Apulia, Corigliano, and Monte Sardo-  all of which were reached by sea. 

Along these routes, the Genoese and the Venetian ships regularly carried crossbowmen, and in several centres –  such as Crete, the Morea in Greece, and in the Cyclades where the Duchy of the Archipelago had been established by the piratical Venetian named Marco Sanudo – there were Latin-ruled territories.  Sanudo’s action is politely described as ‘an independent venture’, and took three years to accomplish. (1207-10).

Here, I should like to refer to one a late-stratum image, used to fill the centre of the Voynich manuscript’s month-diagram for ‘December'(f.73v) .

Archer f73v

I daresay none of my present readers will know, but I published a detailed analysis and commentary on this item among many others some years ago (several years before the version put up by JK Petersen in his blog, or the material posted to Steven Bax’ blog). 

My work remained online until 2017, but my conclusion was not supportive of the often-repeated idea that this figure represents any German or central European figure. I had concluded, rather, that it was intended as an allusion to what was, in medieval times, the popular character for the marker constellation, given its present form here by reference to the type of the marine cross-bowmen who were carried on all Genoese and Venetian ships, including trading vessels.

The published study included a detailed analysis of bow, a point-by-point discussion of Jen Sensfelder’s cautious paper of 2003, and treated the figure’s costume.  It also sought out the earliest appearance of this depiction of Arcitenens (according to Manilius, elsewhere, Sagittarius) as a standing human archer – a task not previously undertaken – and found those origins in the region of  Lake Tiberius, from which glass tesserae, as well as glass workers were evidently imported to assist with the creation of the then-new Opus francigenum (later mis-called ‘gothic’ architecture). It is in early glass windows of that type that our earliest remaining examples of the ‘standing archer’ are to be found in the west.  I note that although no Voynich writer had looked into the question before, nor connected the Beit Alpha mosaic or the Braisne abbey glass with the Voynich figure, since then those illustrations have appeared, without much reason given, in other Voynich blogs and sites.  The historical background and commentary, including the critical matter of translation from the eastern Mediterranean was absent – as of course was mention of my name or the detailed published research which should have made that duplication unnecessary.  

The archer’s costume I read as being composed to create a peculiar, but telling, combination of Aegean Greek and Latin costume, the hat with its turned back brim being recorded both early and late in Spain, and to Spanish marines we also owe the only two surviving examples of that smaller wooden crossbow with the double-lock that explains the depiction of a wooden crossbow together with the curious position of the Voynich figure’s right hand. Unfortunately those two surviving examples date from 1510.

The key to reading that Voynich image is awareness of the constellation’s character in popular lore and in classical texts known to both the eastern Mediterranean and the Latin west in medieval times.  Its character was that of the ‘beast’, the bow-holder (Arcitenens/Saggitifer) – a monster:

Mark where on the ecliptic line the Archer stands,
With outstretch’d bow and arrow in his hands.
When from the east his monster form he rears,

and its rising meant that ships must flee to harbour when he began to raise his bow.

E’en while the sun in Sagittarius lies,
Trust not the faithless sea and cloudless skies. – Aratus 


[300] But even in the previous month, storm-tossed sea, when the Sun scorches the Bow and the Wielder of the Bow [Arcitenens], trust no longer in the night but put to shore in the evening. Of that season and that month let the rising of the Scorpion at the close of night be a sign to thee. For verily his great Bow does the Bowman draw close by the Scorpion’s sting, (Loeb edition).

The scorpion‘ was another form of projectile weapon, seen atop towers in classical images. Mentioned in Roman classical sources, the medieval centuries saw its type maintained in the Byzantine empire, but quite forgotten in western Europe – presumably until the Latins’ gained closer contact with the east during and after the Crusades.

We may associate the ‘holder of the bow’ (Arcitenens), more exactly, with that part of the Aegean taken by Marco Sanudo in his piratical invasion of Naxos, after which it became the Duchy of the Archipelago,  a tiny but ancient town named  Despotikó (of the lords) being found in the Greek archipelago.

The cross-bow wielding maritime ‘lords’ as proverbial embodiment of the ‘monster’ and way-Frankish tower Mytikabarring ‘Sagitifer’ became a very widely-known type – so   widely known that crossbowmen are called not ballistera in the English rolls of Calais, but ‘Saggitario’,  and as late as 1603, Shakespeare knows the Arsenal of Venice as the ‘Saggittary’, the constellation being so named in Flamsteed’s Atlas (published posthumously in 1729).

A Spanish contemporary of Shakespeare, Cervantes also assumes the audience will be entirely familiar with the motif – as familiar as with any proverbial phrase, for he writes in Chapter 44 of  Don Quixote

” … up until yesterday I governed [the island] at my pleasure, like a saggitarius, but … it seemed to me a dangerous trade, that of governor… ” 

Cervantes, Don Quixote.

As so often, historical awareness may be more helpful to an understanding of a draughtsman’s intention than is leafing through digitised manuscripts hoping to ‘find a match’ to suit a theory.  This is especially so if the method of search requires an image to be defined by a subjective choice of a single object from the image, or if the purpose is not so much to classify a manuscript as to learn to read its content.

Even the month-folios’ central emblems, which are legible in the conventions of Mediterranean art, are in the minority within this manuscript whose forms and uses argue different first enunciation for the content than for its manufacture.

I might add -in case you’re interested – that the archer’s hat with its turned-back brim may be attested beyond Spain, but those are where I found the earliest and latest images.  In one dated to the sixth century AD, we see the sort of knitted cap whose form has scarcely changed in millennia, except that fishermen no longer have caps with tails lengthened like this to serve as neck-warmers.  In the second example of those shown below we see a sophisticated version of the same sort of headwear on a character for Japheth, the son of Noah who settled southern Europe after the flood, The detail shows how he is depicted in a fifteenth century Flemish painting. The painter seems to have imagined Japheth entering Europe through Spain, like the Arab armies.   Japheth is shown as a middle eastern character, and in fact the first example is meant to represent either Christ or a Samaritan, according to the curators.   In any case Japheth was middle eastern character like his father Noah, even if the Biblical scheme then has each of Noah’s children found the different ‘races’ to repopulate the world.

It is the hat’s turned-back or rolled-back  brim which is the telling detail here. Note also that the Voynich figure, like that for Japheth, is given a long ‘flat’ face and pointed beard, quite unlike the visual code for a northerner.

costume headwear detail from sixth-century Roman relief in Toledo, where the wearer is meant for Christ or for a Samaritanjapheth-representing-southern-europe-15thc-flemishcostume headwear Arcitenans turned back brim

Similarly the costume is not formed as are Latin medieval costumes, but has a double-flounced skirt à la Grecque, and which may be explained by considering not only modern traditional costume for the Greeks, but certain ceramics from Corinth and the eastern Greek regions made in the 12th and 13thC, one example included here (below left).  Interestingly, another such find  from a Greek speaking centre of that period shows an attempt to imitate the Asian three-colour glaze known as sancai, of which technique I  find evidence also in drawings from the Voynich manuscript’s ‘leaf and root’ section – another of the great many instances where the manuscript announces that its reference – in that case the plants – is not to the Latin’s textual traditions.  Can you see the ‘double flounce’ for the skirt in the enlargement below?

costume skirt 12thc-corinth archer fustanella stylefol-73v-newscan-archers-clothing and bow

I won’t repeat my detailed discussion of the bow, or explain again why I consider the ‘archer’ image more likely to imply an anti-Venetian than a pro-Venetian sentiment. I published that research online and it remained available for other students of the manuscript until 2017, so if you find no mention made of it in any current Voynich site, you may at least find some of my illustrations, albeit re-used in a way which might mislead an unwary reader into supposing they were first found by the re-user.

As I pointed when first summarising the ‘archer-figure’ research at voynichimagery, Despotikó (to find Despotikó follow the line for 25°E on the map below) occupies a place close to the centre of the Cyclades and, in relation to the month-folios which I have always thought more likely to refer to chorographic astronomy than to chorographic astrology, the Greek term ‘chora’ is evocative of ‘Hora’, so that a natural progression runs Horae/[Huri]/Hora/Chora.. and so by association, not by formal etymology, khoros. khorde and korai. I reproduce illustrations from research articles posted through voynichimagery.

Nàxos and Despotikos

map Aegean Cyclades Hora

I also quoted  from the wiki article ‘Despotikó’ which I’ll repeat here to show relevance for to the medieval period (the Latins are here also described generically, as ‘Franks’).

“Currently, excavations are taking place in the northwest part of the island ..The excavations proved the existence of an important late Archaic sanctuary with abundant objects indicating links to mainland Greece, the Eastern Mediterranean and even to Northern Africa, as well as the continued use of this area in the Classical, Hellenistic, Roman and Frankish periods.”

Looking back towards the Mediterranean’s south-west, one recalls that in 1932, Erwin Panofsky spent two hours with the manuscript (not with the usual  rotograph ‘photocopies’) after which as Ann Nill reported, “[Panofsky’s] 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. He thinks there is some influence of the Kabbala in it.”

And then, with regard to inscription of the month-names – variously thought to be in Occitan, in Judeo-Catalan, or in Anglo-French (‘anglo-norman), I note Kokin’s comment when discussing scientific learning among fifteenth-century Jews, of Sicily’s  “deep links to Spanish and Provençal culture” as indicated specifically by one scholar’s writing and reconstructed library.

  • Daniel Stein Kokin, ‘Isaac ha- Kohen ‘s Letter to Marco Lippomano: Jewish- Christian Exchange and Arabic Learning in Renaissance Italy’, The Jewish Quarterly Review ,  Vol. 104, No. 2 (SPRING 2014), pp.192-233.

That Jews had scientific, as well as religious or ‘magical’ literature seems to have escaped d’Imperio and the NSA, despite the publication of Moritz Steinschneider‘s great survey in1893, (Die Hebräischen Übersetzungen des Mittelalters und die Juden als Dolmetscher: Ein Beitrag zur Literaturgeschichte des Mittelalters; meistenteils nach Handschriftlichen Quellen).  For the ongoing translation into English, see

  •  Charles H. Manekin, Y. Tzvi Langermann, Hans Hinrich Biesterfeldt (eds.), Moritz Steinschneider. The Hebrew Translations of the Middle Ages and the Jews as Transmitters.  Volume 1 was published as Vol.16 of Amsterdam Studies in Jewish Philosophy, editor Reinier Munk (2013).

sassoon-gemini-ljs-o57-p-125In connection with astronomy, too, the manuscript Sassoon 823, now UPenn LJS 057, has its ‘Gemini’ illustration from a tradition which is not the al-Sufi corpus latinus though the swollen bellies appear to me to reflect derivation the Asian-Persian style, a link also relevant to the Byzantine Greeks’ updating of Ptolemy’s Tables.

[illustration below added July 12th. I must use a secondary source to illustrate the examples in the lower register. British library is offline at the time of writing.]

Al Sufi illustrated Gemini comparison for blog

To make clear how that connection relates to transition of astronomical matter, and so take the line directly back to Spain (where the Sassoon manuscript was made) from the Black Sea’s eastern side via Trebizond, and thus show why the Voynich manuscript’s atypical quires, and more particularly the septenion might easily have be copied from an original on paper, I’ll now quote a fairly long paragraph.

This nicely demonstrates how Jewish and Byzantine learning passed in tandem, back and forth, along that line between the Black Sea, though the Aegean islands  and Provencal-speaking regions of France to as far as Spain during the thirteenth and fourteenth centuries.

I’m quoting from the Introduction to an important collection of chapter-length essays, including one by Alain Touwaide, an eminent scholar in Byzantine and Islamic medicine, dispensaries and hospitals who was once asked to comment on the Voynich manuscript. Towaide’s paper is now out of  print, but one of his comments was that the manuscript’s binding looked Italian to him, and another that the manuscript’s content recalled the style of Byzantine works of iatrosophia, the sort of hospital handbook and dispensatory, versions of which might contain, in various proportions in various examples, Byzantine medicine and magic.

quote from Lazaris Byzantine astronomy 13t-14thC

  • Stavros Lazaris, Introduction to the chapter-long essays in  A Companion to Byzantine Science (Brill: 2020).

I think that’s quite enough to give you the general picture of the Genoese and Venetian maritime routes, and why they are – as I concluded from research undertaken – directly relevant to the evolution of content now in Beinecke MS 408.

Those who care to study the sort of magical lore found in areas along that line shouldn’t have too much difficulty, I add the following without further comment.

  • Nicholas G. Round, Five Magicians, or the Uses of Literacy’, The Modern Language Review, Vol. 64, No. 4 (Oct., 1969), pp. 793-805.
  • Veronica Menaldi, “Enchanting Go-Betweens: Mediated Love Magic in the Libro de buen amor and Iberian Grimoires,” in Ryan D. Giles and José Manuel Hidalgo (eds.), A New Companion to the Libro de buen amor. (Brill, 2021) pp. 75-88.

amulet Jewish and scripts Salonika 17thC

The Voynich script has no ‘X’ shape glyph.


Additional notes [added 13th July]

  1. There are two sites named Despotikó in the Cyclades, the other – better known today – on Mykonos.
  2. Paragraphs inadvertently omitted after discussion of the archer’s hat, and his ‘Spanish-Arab’ face and beard, included the fact that in 1317, the Duchy of the Archipelago had been raided by the Catalan company.  In this note, I’ll just quote from the wiki, rather than from the sources used in my own work because this is only to illustrate historical connection between the Duchy, Venice, the Spanish marines, Constantinople and Trebizond.

“The Catalan Company; or the Great Catalan Company’ (Catalan: Gran Companyia Catalana, Latin: Exercitus francorum, Societatis exercitus catalanorum, Societatis cathalanorum, Magna Societas Catalanorum) was a company of mercenaries led by Roger de Flor in the early 14th century and hired by the Byzantine Emperor Andronikos II Palaiologos to combat the increasing power of the Anatolian beyliks. It was formed by almogavar veterans of the War of the Sicilian Vespers, who had remained unemployed after the signing in 1302 of the Peace of Caltabellotta between the Crown of Aragon and the French dynasty of the Angevins….”

In 1248, the Duchy had been nominally granted to William of Villehardouin, Prince of Achaea. Marco II Sanudo lost many of the islands, except Naxos and Paros, to the forces of the renewed Byzantine Empire under the admiral Licario in the late 13th century. The Byzantine revival was to prove short-lived though, as they relinquished control of their gains in 1310.

In 1317 the Catalan Company raided the remnants of the Duchy; in 1383, the Crispo family led an armed insurrection and overthrew Sanudo’s heirs as Dukes of Archipelago. Under the Crispo dukes, social order and agriculture decayed, and piracy became dominant.

The figure for the archer as holder of the ‘arc’ is among the many which eventually led me to date the last phase of the Voynich images’ evolution to no later than 1340 – barring the usual exception of late-added pigments, post-production marginalia etc. and – possibly but not necessarily – the ‘Mongol-dressed preacher” diagram.

3. Also in this connection, the type of Greek skirt given the archer is related to the Dalmatian ‘fustanella’ widely adopted elsewhere, especially under the Turks, and I see today that the wiki article ‘Fustanella’ refers to the same scholarly study, and includes the same illustration which I had from that source). The earliest remaining example from Dalmatia is a small statuette roughly contemporary with the first of my Spanish examples for the ‘fisherman’s hat’.i.e. 6thC AD.  We note also a type of Venetian galley was known as the ‘fusta’, whose date of introduction is unknown, but the few documentary references which have been found so far come from fifteenth-century records.  My chief reference here is Royal & McManamon though their article – for obvious reasons – is focused on the period post- 1450. 

The term fusta is of Italian derivation, and Venetian manuscripts from the fifteenth and sixteenth centuries supply skeletal information on the vessel type. In the fifteenth century, fuste had 10-26 rowers’ benches and a length on deck from ca. 16.5-33.0 m, the largest of which was only slightly smaller than a Venetian light galley (Zorzi da Modon, fols. 27v-28v, 37v-39; Anderson 1925, pp. 145-147; Chiggiato 1987, p. lxix). By the sixteenth century, fuste were more regular in size as they had 18-23 benches. An anonymous shipwright writing after 1546 noted that a fusta of 20 benches was almost the same size as a light galley, carried ordnance in contrast to a bregantin or fregata, and had a draft of 0.87 m once armed (Pre’ Teodoro, fols. 14—15, 35v; Tucci 1963/4, pp. 282-283; Picheroni della Mirándola, fol.7).

  • Jeffrey G. Royal and John M. McManamon, ‘Three Renaissance Wrecks from Turkey and Their Implications for Maritime History in the Eastern Mediterranean’, Journal of Maritime Archaeology, December 2009, Vol. 4, No. 2 (December 2009), pp. 103-129. (p.106).