Once we abandon the hope that the Voynich text may be a natural language, or an encryption of the same, and approach it as a system of notation, many aspects of the text become illuminated. Many things about the text become less puzzling and more understandable.
In particular, the highly repetitive nature of the text is immediately explicable. If the text is some sort of record - measures, or readings - made in a system of notation, then we would expect repetition.
Ordinarily, such information would be presented in a table form, and we expect that entries in tables are similar. Measures of rainfall, for instance, might all be similar, with repetition, across a set of tables in a single season.
In our manuscript, by my current proposal, this sort of data has been presented in a way that resembles a written text.
Nevertheless, we can see tabular-like repetition throughout the text, especially in the context of LINES.
Lines, it was long ago observed, seem to behave as "functional units". A LINE, that is, is like an ENTRY in the record, or records over a set duration, and a whole unit in itself.
Very often, in lines, we have a series of words that are all similar. Here is a short line from f2, for example:
sho.cholo.cheor.chodaiin
Very often, lines rehearse variants of a particular configuration of glyphs. Here is another example:
dor.chol.chor.chol.keol.chy.chtydaiin.otchor.chan
This line rehearses [ch] words, all slightly different. And sometimes there is serial repetition of whole words:
qotcheaiin.dchain.cthain.daiin.daiin.cthain.qotaiin
What sort of text behaves like this?
The answer we can supply, is: a record of astrological events, or coordinates, or similar data, which has been compiled into lines rather than set out into tables.
Otherwise, we struggle to provide any natural reason for a text such as this, or any account of how a natural plaintext could have ended up in such a form.
* * *
The history of the study of Voynichese is a history of frustration. The amount of expert brain-power - and more recently computational power - applied to the mystery has been prodigious.
In almost all studies there has been the assumption (acknowledged or not) that there is a natural language plaintext to be recovered. How has it been manipulated, encrypted, scrambled?
There are countless possibilities. Is the text heavily abbreviated? Is it just consonants? Is it a phonetic rendering of a foreign language? Is it a verbose cypher in which most of the words are nulls?
But it is hard to explain any process that would result in such a display of repetition. Linguists are left clutching at straws.
Famously, Stolfi - who conducted the most in-depth analysis of the structures of the words - thought the behavior of the text resembled Chinese characters more than a semantic text in a European language - even though that is what it looks like.
Why the text is not what it looks like is an on-going conundrum.
It looks like a meaningful, flowing text of words that can be read and pronounced. It looks as though its purpose is to explain or provide commentary upon, or at least accompany, the illustrations.
But on closer inspection, this prima facie impression is problematic. There are many critical ways in which the text is not text-like at all - the extraordinary amount of repetition being one of them.
Readers of this page will probably be well aware of this phenomenon, but let us underline it again. The text is extraordinarily repetitive:
qokshedy.chedy.qokedy.chkedy.daiin.shetar.shedy.qekaiin.chedy.dsheey.qotaiin.checkhy.qoty.cheg.shedy.qokeey.rchedy.qokeedy.lo.schedy.chedchy.qokal.olchedy.qokaiin.chedy.qokeedy.lchedy.qoky.solshed.lsheedy.qeeedy.qoky.oqol.rsheedy.qokedy.qoteedy.qoteedy
What typifies the text are series of similiar words (units), all variants of the same configuration of glyphs, all slightly different, modified with a system of prefixes and suffixes.
[qoteedy], [qokedy], [qokeedy], [qokshedy] and so on are all modifications of the same word.
Thus we have lines of text consisting of a series of slight modifications of a single word. That is the nature of the repetition.
On reflection - and with the utmost sobriety - it is clear that there is unlikely to be any linguistic explanation for this (without invoking quite non-linguistic processes to account for it.)
* * *
This gives rise, then, to non-linguistic proposals. The work of Timm and others has demonstrated just how deeply non-linguistic the text is, its repetitive nature being a prime characteristic and itself strongly suggesting some process of auto-generation.
The text is strikingly mechanical and artificial. Timm's study emphasized that whole passages, often adjacent, are duplicated with minor variations.
His explanation for this is what he calls 'self-citation'. It seems as though the scribe is copying an earlier passage, citing himself, as it were, and making minor changes as he goes.
Timm concludes that the text is an elaborate hoax, variations on gibberish.
It is a 'lazy scribe' scenario. It is actually arduous to compose a long stream of fresh gibberish. It is easier to look back at the gibberish you've already written and keep altering it.
Thus, Timm says, the highly repetitive text.
But this too assumes a natural language plaintext in the sense that the text is supposed to imply one, the hoax invokes the illusion of one.
Timm considers and rejects other scenarios but argues for 'self-citation'.
His self-citation hypothesis, however, over-estimates the liberty of the scribe(s). If a scribe is copying an earlier passsage, they are under severe constraints as to what variations they can make. Not only is the text repetitive, it is under the constraints of a strict set of rules governing modifications. The variants are not random or made at a scribe's whim.
The self-citation phenomenon, though - namely similar but slightly different passages near to each other - is well-explained if the text is a record of astrological or similar types of readings.
To use the example of rainfall again (because there are clearly depictions of rain cycles in the diagrams), it is entirely natural that there might be adjacent periods of similar but slightly different rainfall, or weather in general.
If the text records the ever-changing moods of natural cycles - celestial or terrestrial or both - there is no need for a 'self-citation' hypothesis, and the text is not repetitive gibberish. Rather it is numbers, measures, a record in a system of (astrological) notation. Repetition and similar passages are natural to such records. Such records are, of course, meaningful texts, but not semantic ones.
* * *
We can say 'astrological notation' because that is the nature of the manuscript. It is a late medieval cosmological herbal. It conspicuously concerns celestial (especially zodiacal) cycles, and corresponding meteorological and calendrical cycles by extension.
Exactly what cycles are recorded is another matter, but in the first instance the notation system appears astrological, cosmological, calenderical, mapping for example the quarters of the year and the movements of the sun.
That appears to be the main framework, established in the system of the so-called gallows glyphs which are surely at the heart of the text.
But the YEAR is itself a profoundly multivalent symbol. Its patterns are isomorphic. The astrological cosmos is replete in rich symbolic associations that have an almost geometric coherence.
The winter solstice is, at the same time, the cusp of Capricorn, the house of Saturn, cardinal Earth, the direction south and midday on the clock.
In a system of astrological notation, one glyph might carry all of these parallelisms and be used in multiple ways or suitably modified.
It is inherent to such isomorphism that it will create patterns of sameness and difference.
Variants like [chol], [chor], [chom], [cheor], [shol] might all refer to the same thing but at different levels, a different order of parallels. For example something like:
[chol] = spring, [chor] = dawn, [chom] = Aries [cheor] = cardinal Fire, [shol] = east.
The Voynich system is probably not excessively complex. In essence it is probably simple. But it will be as simple, and complex, as astrological symbolism.
* * *
One of the most curious things about Voynichese is that it seems to have only been deployed once and never appears again.
If it is a system of astrological notation - and a system necessarily of some sophistication, and the product of both labour and invention - why was it never used again?
Given the unpolished and somewhat rustic nature of the manuscript, it would seem to be a private project.
It is far from being a professional manuscript, although it affects the style and presentation of more professional works.
The conclusion to draw is that Voynichese is a private system of notation, developed for the keeping of private records. In practice, then, - plausibly - it is like a system of astrological shorthand that the author has developed for his own use and utility.
It is thus idiosyncretic, but its utility to the author and its basis in natural cycles ensures its intelligibility.
* * *
One of the things that might count against a 'de-tabulated' text is that the manuscript is unruled. This is part of its rustic, unpolished character.
The lines of text seem to be made by eye and, in fact, some pages are an outright mess.
Not only are there no tables, there is no disciplined attempt at a properly ruled, spaced, aligned text at all.
If a book of tables is the plaintext, not only have the tables been removed but most of the formatting has been removed as well.
There is little left visually of a tabulated text, although there are visual features that might be consistent with it, such as certain glyphs preferring the top lines of paragraphs and even cases where the text seems to have been written onto the page in vertical sequences.
In other places it seems that an upper line leaves space for glyphs in a lower line raising questions about the order in which the lines were placed upon the page.
The appearance of a normal text written line by line, left to right, is not altogether true on closer inspection.
Arguably, there are vestiges of tables, patterns suggestive of underlying tabulation, to be found.
* * *
The lack of tabulated data in the Voynich ms. deserves notice but, from my wider reading, has rarely been mentioned over the years.
There are no tables in the manuscript. Data is set out in many ways. Circles, especially. But there are no tables, either with lines or without.
Tables were, though, a popular format in cosmological manuscripts. Astrological data lends itself to tabular presentation. Why has no information been set out in tables of any kind in the Voynich? The nearest is the text in Quire 20 which seems to have 'catalogue' format.
The answer might be that the running text, set out in lines and paragraphs, is inherently tabulated information (but without the tables.)
This is not to say there was a plaintext with data set out in neat tables. Rather, it seems our author was not in the habit of keeping his records in a table form, but rather as (text-like) notation.
* * *
This ingenius system of notation is suggested to our author, I propose, by the alphanumeric tables of the Greek edition of Ptolemy's Canones.
Latin is not alphanumeric (Roman numerals notwithstanding.) But all the other scholarly languages of the era were: Greek, Hebrew, Arabic.
In the Canones (Handy Tables) we see (elegantly presented) what seem - to the Latin man - like short words or bigrams or syllables arranged in tables. But they are, of course, numbers (as he no doubt knows.)
The Canones are lists of solar, lunar and planetary positions, and the increments necessary to quickly calculate their actual positions. 'Handy Tables' indeed. The tables are set out in Greek numeration: Greek letters as numbers.
Nevertheless, I propose that our author is impressed by the notion that celestial movements produce a text.
The right glyphs applied in the right way will turn a record of celestial or natural phenomena into a (quasi-linguistic) text.
It is assuredly repetitive, with a limited set of glyphs and bigrams and possible combinations, but it resembles a text all the same. Readings look like words and rows of readings look like sentences.
In the simplest analogy, records presented in table form in alphanumeric Greek might be taken from the tables and arranged in lines instead. A row in the table becomes a line of text. The cells of the table become the word breaks.
Thus, the lines of Voynich text have been de-tabulated. When the repetition is framed by a table nothing seems untoward. But if we remove the table it looks like a sentence with an unreasonable amount of repetition.
It is the repetition - repeated series of variants - that I am attempting to explain here.
There are sundry problems with this simple analogy - it cannot be quite that simple - but it still offers a plausible genesis for the Voynich text.
It has been suggested (many times) before, I know. Is the text detabulated data? It remains a better explanation for the repetitive nature of the text than any natural language or encryption scenario on offer.
* * *
There is much else, I maintain, connecting the manuscript with Ptolemy's Canones (Vaticanus graecus 1291) but I propose that, in the first instance, it provided the inspiration for a system of notation that is quasi-linguistic.
In any case - and this is often overlooked, especially by researchers using EVA - Voynichese is alphanumeric in an obvious sense. The glyph set is made of letters and numerals.
The notation system might not be alphanumeric like Greek (although it might be?) but the glyph set is inherently alphanumeric in its composition and suggests an alphanumeric model as its inspiration.
The glyph set is eclectic. Why does it include letters and numerals?
An 'astrological notation' proposal also solves the question: where are the numbers? If the text accompanies the illustrations, surely there must be numbers?
It is inconceivable that all those circles and star charts do not include any numbers. Are they all written out in full?
In Voynichese, there is apparently no distinct numeration system separate from the 'alphabet'.
Again: why do we have a text in which letters and numbers are mixed together and then presented like an 'alphabet'?
The glyphs [q], [d] and [y] are of outstanding significance in the text (especially Text B), are copious, and are all numerals.
We can explain this by proposing that the text is a system of notation that is essentially numeric, being cosmological and based upon, embedding, cosmic cycles.
It has sometimes been suggested by seasoned researchers that glyph series like [eee] and [iin] might be numerations, perhaps working not unlike Roman numerals.
It is a non-linguistic possibility that has not been fully explored. (Or, more likely, it has been thoroughly explored in some backwater of Voynich Studies and no one took any notice.)
* * *
There is deep and quite understandable resistance to the idea that the text is a record of astrological data, or meteorological data, or similar. I feel this resistance myself.
The text promises so much more.
Have we been labouring for endless hours over a text that is a record of moon phases and rainfall?
Does it not contain any decent semantic information? No wisdom? No health advice? No account of women's diseases? No account of the powers of herbs? No theological reflections? No lost herbal text from ancient Greece?
Astrological tables is a disappointing outcome. It is something much less than what so many had hoped.
All the same, that is where my investigations have led me, come what may. It appears to be the most adequate explanation, given the evidence, internal and external, textual and contextual, in toto.
From the outset, the unnaturally repetitive characteristics of the text have been striking. And so have the contortions necessary for researchers to construe that phenomenon as linguistic, encrypted or not.
There is a more straightforward explanation. What has been recorded in a form that looks like a linguistic text are records made with (quite a small) set of notation symbols, of astrological significance in the first instance but probably concerning corresponding natural cycles as well.
The text is not empty, as hoax scenarios propose, but neither is it full of the semantic content suggested by the textual presentation.
If we can decipher the notation it is likely the Voynich text will be as exciting to read as Ptolemy's 'Handy Tables'.
R.B.
No comments:
Post a Comment