O’Donovan notes #6b – Weighing images; the analytical-critical approach.

c.3200 words

The author’s rights are asserted.

At some level everyone knows, or at least remembers, that their first response to the drawings in this manuscript was simple bewilderment. They made no sense; you couldn’t read them.

In response to that immediate feeling of disorientation you might have started skimming past dozens of drawings and dozens of pages seeking something, one image or one detail, that did seem to make sense – to chime with something already familiar. As I’ve said before, this is an instinctive response to a strange environment and unfamiliar surroundings; it’s how we can narrow our vision so that we ‘see’ none but the one familiar face among the crowds at a railway station in rush-hour.

We’re hard-wired to recognise the familiar, or seemingly-familiar.

To an extent, this is central to the cryptographer’s work, too. The difference is that cryptographers pay just a much attention to differences, where in Voynich studies those using the images as support for a theory don’t.

Another reaction against that first sense of bewilderment is to say to ourselves the equivalent of ‘Hush, dear, it’s alright. It’s just Dad in a Santa suit.’ We imagine that the unreadable *is* really normal underneath it all and never pause to examine whether our idea of ‘normal’ isn’t pre-judged. That presumption of the normal as western Christian European was the Friedmans’ flaw. It’s still embedded in the traditionalists’ approach.

A third reaction is to blame the object for being as it is, and for not being more comfortable and familiar. It’s not our fault the pictures bewilder us, we say. Hence the earlier invention of a lone fictional character called ‘the artist’ on whose imaginary head was piled blame and various calumnies – he was a child said some, a child-genius said others, a mad-man asserted another, a deeply devious individual out to mislead us… etc.etc.

By means of one – sometimes all – of these defensive reactions, that initial true recognition that these images are NOT much like forms seen in medieval Latin (i.e. western Christian) manuscripts was covered over, dispelled, rationalised or just ignored.

Theories also served to reduce feelings of bewilderment. In creating a theory, an individual felt once more in control. The all-at-sea feeling passed; rationalisation replaced enquiry and reason was quite unconsciously replaced by mere logic, operating from premises which were untested, doubtful, historically inaccurate or otherwise less than useful.

That’s what happened with Mary d’Imperio, as she persuaded herself that the images “of extreme oddity, quaintness. and foreignness …” were really just quirky versions of what she’d expected the manuscript should be, underneath it all: an ordinary western Christian composition. Its plants should find matches in the Latin herbals and so on.

The images didn’t change; she had suppressed her immediate perceptions and thereafter saw the images clearly no longer, but through the screen provided by the groups’ fixed assumptions.

d’Imperio was not stupid; she was a highly intelligent person who prided herself on an ability to think coolly and clearly. Perhaps the kindest thing to say is that she had no appropriate prior studies, didn’t appreciate the vital importance of stylistics, and never went outside the cryptographers’ group to get better information. She couldn’t. The study was classified by the NSA.

The sweep of human history shows that almost everything human beings do is done for the sake of comfort – physical, emotional or spiritual- so we can hardly blame others or ourselves for being inclined towards the comfortable and familiar. Yet since, at some level, we cannot forget our first perception that these images aren’t nice and normal products of European culture, the first critical question to ask is – what am I presuming is ‘normal’? and secondly, why -exactly – are these images not ‘normal’ in those terms?

Bad artist?

Let’s start with the old story that the ‘artist’ was incompetent. (pause: what evidence is there for supposing these are original creations and not ones copied? What evidence is there for supposing all were made by just one person? ..).

Instead of trying to run away from the ‘I-can’t-understand’ sort of problem, we move close up and cross-examine it.

Using the same example from folio 79v, we ask about it- Exactly what about this image is making it unreadable for me? What, exactly seems ‘wrong’? If you assume a European norm – what is it that differs so much from the conventions of Europe’s medieval art that I while I can read those pictures pretty easily (Latin language text or not) I’m struggling with this one?

(Form, disposition, stylistics, spoken tongue, cultural-artistic conventions and subject-matter usually lie at the heart of such problems, but since this is a first exercise, I’ll take things easier. For the same reason and though one must, as a rule, comment on the whole of any image, I’ll just speak about this female figure rather than have a post three times as long,).

Exactly why doesn’t it/she look like a figure from a medieval European manuscript made earlier than 1435?

Again, to answer that question as a professional would, you’d have to locate this image against the wider landscape of historical and cross-cultural studies, and read papers having such titles as, ‘Attitudes to the human body in Europe’s medieval art prior to 1440 AD’ – but for the exercise, assume that’s been done.

No harm in testing the validity of a previous idea, so we start with the idea earlier common that the ‘artist’ was incompetent. Does the primary evidence agree?

Look at the finer details – the way the hair is shown as being swept back from the head, the hint of shadow included to distinguish the line of the head-veil from the hair. If you consider the face alone, you see that the spacing of eye-to-eyebrow, and these to the top of the nose and mouth are remarkably good. I say ‘remarkably’ because the face measures no more than 5mm square (yes, five millimeters). That’s 13/64ths (0.2) of an inch each way.

You find that level of practiced precision and hand-eye co-ordination in people whose work is cartography, gem-engraving, ivory-carving and a few other trades. It’s not necessarily found in every manuscript illuminator. Jewish artists in micrography (extremely small-written letters that are formed to create the impression of a drawn image) also had the necessary practice and skill.

Some years ago I asked the advice of an eminent specialist in Hebrew palaeography if the micrographic letters on some petals of the violas (folio 9v) made any sense. The specialist’s opinion was that they did not spell out words, but read as if an untrained person had attempted to copy Hebrew micrography. The individual letters were intelligible, and the specialist read them off, but remarked that the forms were untutored so that e.g. vav and yod were not clearly distinguished as they would be by a Hebrew scribe and the string(s) as they appear there make no sense. One of the same illustrations I used in those posts happens to be included in Koen Gheuen’s recent post about ‘Spirals’; additional examples can be found in the very poor wiki article ‘Micrography’.

Next, we consider the lines which form the figure’s torso.

Can you see what firm, sure, skillful lines they are which shape the curves of the figure’s back and belly so economically?

See how the figure’s right armpit is indicated by a single, practiced, effortless line met by that which gives – in good proportion – the line of the right breast, and the other line (which no unskilled draughtsman would include) which extends the line of inner right leg to suggest the small hollow between hip and belly?

The combination of effortlessness and accuracy at such a minute scale leaves little room to doubt that this fifteenth-century draftsman was highly competent and practiced – and we may add to those markers just noticed the relative proportions given the torso and lower body down to just above the knee. So why not draw the whole figure ‘normally’? The most obvious possibilities are that (i) what we are seeing here reflects non-European custom (ii) the draughtsman didn’t want to make it all ‘normal’ (ii) the person paying for the materials didn’t want that. Most medieval manuscripts are copies of older works.

However, comparing this figure to others in the month-folios and ‘bathy-‘ section brings to notice two interesting points about this figure’s right arm.

First, that the fifteenth-century draughtsman seems to have forgotten, in drawing that limb, that these figures were supposed to have broken- or boneless limbs arms (i.e. deprived of strength in the idiom of certain languages), and he has drawn it with its proportions pretty right, and evoking an underlying bone structure. More remarkable still for this manuscript, if it’s intentional and not co-incidental, is that the figure’s right hand seems – almost – perhaps – to be drawn foreshortened!

One can’t be sure of the last; there’s not enough detail to be sure, but given the general absence from these drawings of any hint of that movement towards literalism or the styles for rendering perspective which mark Europe’s later medieval art, even a hint of possible foreshortening is worth mention.

Fixing things.

If we were to isolate this figure, give it/her another head, fix her left arm to look more like the right arm.. we might end up with something more nearly compatible with art of Italy, France or Spain between c.1350-1430 AD. The ‘swollen belly’ came via strands of imported astronomical imagery, to appear in art of the Latin west as a marker of celestial/heavenly ‘bodies’ (not, initially, related to the planets). In that way we start to see it in a couple of medical texts, or (as Ellie and I happened to notice at much the same time) in illustrations produced in regions under English or French control during the early 1400s, after which it soon became a fad in western art. So the swollen belly isn’t a problem even for those having all-Latin theories. But to make the whole figure suit the theory, the limbs and the head would have to be ‘fixed’ rather than be paid attention.

And that’s exactly what so many Voynich writers do.

The Voynich manuscript’s drawings have constantly been treated as if their form and stylistics could be arbitrarily ‘fixed up’ the better to suit a theory. The theory says they ‘ought to be…’ Aztec, or European or whatever, so the theorist produces comparisons that are Aztec or whatever as if the two were close equivalents.

Most of the writers who do that do not seem, to me, to be conscious of what they’re doing. Rather, they appear to just imagine/filter the Voynich manuscript’s images so they see them as if images first given form in the environment the theorist prefers.

The usual technique that you’ll see employed to encourage others’ belief in a theory is the presentation of alleged ‘matches’ – pairings of a detail from the manuscript with one selected from within parameters demanded by the theory, and very often with no effort whatever made to treat of the Voynich drawing itself. Commentary thus becomes a commentary on the theory, not on the manuscript. All the important aspects of art-commentary such as date and place of making; origins; stylistics and so on, are ignored or presumed covered by the theory.

That’s how it has been, almost without exception, since the end of that brief period when Jim Reeds mailing list for Friends of the Voynich manuscript was marked by a spirit of enquiry rather than theory-formation.

To explain how propagandist-style ‘matches’ work, let me give as example an idea popular for a time, but which hasn’t been revived recently – so this should offend fewer current Voynicheros.

When attempting to explain the unclothed figures around the month-folios and in the ‘bathy-‘ section, those determined on a theory of medieval Latin (i.e. western Christian) European origin for the images had only a limited choice from which to pick their proposed matches, because other than the often salacious images found in copies of that innocent poem ‘Balneis Puteolanis’ there were relatively few unclothed female types in pre-Renaissance Europe – the most obvious being Eve, Magdalen (half-clothed/hair-shirted), and Melusine.

Melusine became a popular choice for a time. In Latin art, she is often depicted in a bath that is barrel- or tub-like and her legend is one native to central Europe and France.

Below is an image of Melusine from a manuscript made around, or after c.1435. Notice the elongated torso, low-slung belly, long nose, elongated and slender neck, defined chin, the network ‘fez’ on her head and the stave-made tub. Not characteristic of the Voynich drawings and none of those details agrees with this ‘match’, does it? – but theorists are generally positivistic in their approach – taking the attitude that we should hunt similarities, and that to mention differences is just spoil-sport behaviour which should not be allowed to interrupt. And some quite honestly cannot see differences at all. Strange but true.

Pairings no less casual than that (above) were, and still are, presented by theorists who believe – and whose sole aim has been to persuade the audience to believe – that their pairings are so similar that you could, in effect, substitute the ‘match’ for the Voynich image itself – thus tacitly claiming the Voynich manuscript’s visual language translated without its being considered.

A theorist doesn’t experience their creating such pairings as a practice misleading, erroneous or illusory. The positive comparison is a product of their theory, and thus supports their theory, so they regard it as a good thing, and in that sense a good match.

Having no experience of what is the usual methodology in the wider world, and no very deep prior study of such matters to inform their ideas, it is not surprising that they find this work fairly easy.

Once a given theorist believes they’ve found a couple of matches between their theory and the Voynich pictures, they’re past the tipping point. I say this from years of constant association with the online ‘Voynich world’.

Once, it was expected that any theory about the drawings would convert into a valid theory about the written text, but after so many years during which theory-driven narratives have proliferated and been elaborated, the two areas of Voynich studies are now pretty much divorced. Objective standards and rigor are expected by one but actively opposed by the other. To discuss methodology in treating the written text is accepted in any Voynich arena; to so much as invite interested persons to discuss it in relation to historical-pictorial narratives has been, quite literally, prohibited. Not because it had caused dissent at any previous time, but because if people weren’t prevented from talking about it, some theorists might feel upset. Nobody got to vote, and the manuscript never gets to.

For anyone interested in the images and in taking a different approach, here are two vital principles.

Meaning is context-dependent. Differences really matter.

Let’s put that image of Melusine back in its proper context.

Look at the differences between that page and folio 79v from the Voynich manuscript.

Consider the Melusine page’s layout and its having ruling-out; the script’s strong and emphasised verticals, the illustration’s attempt at literalism, the costumes, the way those head-veils are designed, the effort at perspective, emphasis on furniture, rich clothing and individuals’ relative social standing… all typical of western Christian manuscripts of its time. These things aren’t characteristics of the Voynich manuscript or of its drawings. One has to consider the question, ‘How much has to match before you ascribe the origin of a drawing, or set of drawings, to a certain cultural environment?

(If you’ve never heard of Melusine before, the wiki article will do for basic information. For an illuminated manuscript copy (incomplete) of the fourteenth-century Roman de Mélusine in which she looks, when dressed, as if she had legs – see British Library MS Harley 4418.

One obvious difference between images of Melusine and ‘ladies’ from the Voynich month- and bathy- sections is that Melusine was a hybrid creature whose lower half was dragon-like, sometimes drawn more serpentine, sometimes having clawed feet on a dragon body. Only when dressed could she pass for someone who had legs.

Our figure in folio 79v has legs, even if the shanks look far too thin. The tokens – the markers – which told a medieval Latin audience that a figure was Melusine are not there. So if meant for Melusine, the maker was not part of the Latins’ tradition and if he were, this is not Melusine. The same applies to other figures, including those shown unclothed and in roughly-cylindrical containers around the month-folio’s diagrams.

And now we come to another, and a most important difference between the norms of western medieval art, and what we have in the Voynich manuscript.

Proportions.

In relation to the human body..

Among the set conventions of pre-modern western Christian art is that the proportion of a human figure’s head to its torso shall be literal. The length of your face (hairline to chin) is normally about the same as the distance from the hollow of your throat to about the middle of your chest where the rib-cage meets. Test that out; don’t just believe me.

Medieval draughtsmen typically maintain those proportions, even in drawings small enough to appear in ornament bas de page. Here’s another Melusine, now as mermaid. The mark for the hollow of her throat is set a little low but I think you’ll see what I mean about those proportions – in the European image they are pretty right.

The Voynich figure’s face is much larger than the conventions of medieval Latin art allow – almost double the literal length, for the face is as large as the distance from the throat-hollow to below the position where you’d place the navel.

Koen Gheuens once produced a long and detailed post which concluded that the Voynich figures’ proportions were ok. It was a case argued so very well that I was quite persuaded by it too – until next I looked at the manuscript. I do agree with him that the figures in some of the ‘ladies’ drawings have faces less disproportionate than this example.

Differences matter.

The Voynich figure is unclothed yet has legs. Only the right arm’s proportions accord with Melusine figures in western Christian art. To argue as so many do by these ‘pairings’ that e.g. ‘Melusine=Voynich figure’ may create an illusion sufficient to persuade others of a theory, but cannot do justice to the Voynich drawings, assist those working on the written text, or deal fairly with the manuscript itself.

Knowledge of comparative studies in culture and art provide a better ability to recognise stylistic and cultural indicators, and train a person to take account of both similarities and differences. Such preliminary study enables a person, without any additional information, to know that the image shown below, for example, is no product of Chinese work, but a European painter’s work. The Chinese accept and understand that the surface is two dimensional – they ‘wrote’ paintings. After re-discovery of the techniques of literalism (with a small ‘l’), Europeans fought that two-dimensional reality and tried to have things look moulded, as if they were working with cloth, stone or clay. Look at the leaves this image.

I’m using Asian versus European examples below because they provide comparisons-and-differences which a predominantly western audience should find easy to recognise; not as support for any theory.

In the same way, a knowledge of comparative customs and cultures is how we know that while both the following items were found in Italy and both had been owned by the same person, here again one was made in China and the other in Europe.

The bowl was owned by, and the book written by Ulisse Aldrovandi (1522 -1605) often described as “the moving force behind Bologna’s botanical garden” and it was he who invented a phrase “herbals of the alchemists” for a genre of foreign-influenced European herbals.

Another example: you see (below, left) a figure of the Simurgh/Phoenix whose form and expressed character shows the maker was native to the east, where the type is celebrated in art, literature and popular culture – from Persia to China. By contrast, the other (below, right) is a very nicely-presented and well-rendered form, but one plainly ill-informed, produced by someone from a very different cultural environment. The version on the right is so clearly a European work that an experienced analyst could say so before being provided with any more information. Differences matter because they carry important information about distinctions in time, place and cultural context.

… and you know they do.

So long as you’re dealing with images expressing matters natural to a culture with which you’re comfortable, differences and their significance are apparent to you so easily that they seem self-evident. For example – which of these two self-contained paintings (below) is a religious character, do you think, and which a secular one?

.

And now, at last, the whole point of this post.

How do you know that?

If you didn’t need to go hunting wiki articles to answer the first question you don’t need them to answer this one, either.

Just by carefully, systematically and slowly scanning each picture, noting both similarities and differences, you can bring to your conscious mind the points of difference you processed so fast that you didn’t consciously list them.

How – exactly – did you recognise one as a religious figure and the other as secular?

That done – imagine the opposite opinion, and how an argument between the two might run. You’ll find that the correct opinion is more difficult to explain than you expected.

And that little problem was your first exercise in iconographic analysis.

Cheers.