| Welcome, Guest |
You have to register before you can post on our site.
|
|
|
| previous research on Arabic and the Voynich ms? |
|
Posted by: geoffreycaveney - 18-09-2020, 11:37 PM - Forum: Analysis of the text
- Replies (17)
|
 |
The following is very speculative so I'm moving it out of the "Verbose cipher" thread before I even put it in there.
This also has nothing to do with any of what might be considered my own "pet" theories about the Voynich ms, currently or previously.
But my You are not allowed to view links. Register or Login to view. of the Voynich ms text, and the You are not allowed to view links. Register or Login to view. on folio page f67v1, unexpectedly drew my attention to Arabic for a couple reasons. I wish to emphasize, Arabic would not by any means be my first candidate to be the language of the Voynich ms. Of course it is possible, like many languages are. My first inclination, like Ventris with Linear B and Greek, would be to first try to find evidence to rule out Arabic as a candidate language if possible, and see if such efforts succeed or fail. Ventris was led to the Mycenaean Greek hypothesis as a result of a line of investigation to try to prove that Linear B wasn't Greek, which to his own surprise failed and led to the opposite conclusion.
One of the VCI "Zodiac" labels reads <d'hrá>. I happened to discover that al-eadhra' is the Arabic name for Virgo (for Arabic-speaking Christians, Maryam al-eadhra' means "the Virgin Mary"). Now before anyone gets too excited, I have not been able to find many other such matches among Arabic Zodiac sign names, although the issue is apparently complicated by the fact that several different sets of Arabic Zodiac signs and names appear to exist, and who knows what the situation was like in the late medieval period. An investigation into Arabic in medieval Europe would also have to include the Maltese language, for example. But I digress. Another VCI "Zodiac" label reads <káás>, and the Arabic name for Sagittarius is al-qaws. Alas, this label is four "houses" away from <d'hrá>, while Sagittarius should only be three houses away from Virgo. More vaguely, the VCI Zodiac label <da aly> loosely resembles al-dulu, the Arabic name for Aquarius. At least <da aly> is two houses away from <káás>, the right distance from Sagittarius to Aquarius.
This is a curiosity, but hardly convincing. The best of the name matches, <d'hrá>, is the one in the "wrong" place in relation to the other two roughly similar names. And a quick check has not found any explanation for what the other nine label names might be related to. Of course at this provisional stage many of the phonetic values of the VCI transcription system could be completely wrong, but changes that improve other label names might also disturb the neat match of <d'hrá>. For now it is only a few curious names, which may or may not mean anything.
I must also mention that naturally I had to check the VCI transcription of the two label words near the group of seven stars that appear to represent the Pleiades. In fact I did this before I did the 12 "Zodiac" labels. The VCI readings of the two vords are <khaly> and <hajez>. Hejaz is the historically significant Arabian coastal region along the Red Sea, in which Mecca and Medina are located. Khali is the Arabic name of the large desert bordering Hejaz to the east. How this would be related to the Pleiades, I don't know. One wall of the Kaaba in Mecca is aligned with the rising point of Canopus, the second-brightest star in the night sky after Sirius.
I never delved much into Arabic in relation to the Voynich ms. I always figured that hundreds of millions of people around the world know the language so well already, that I never thought I could contribute much to such an investigation without a great knowledge of the language myself. I also figured if the ms was written in Arabic, surely someone would have figured that out and deciphered it already. But maybe everyone else has always figured the same thing too, so that the hypothesis has perhaps actually not been as deeply investigated as other well-known languages have.
The one tell-tale giveaway sign of the Arabic language should be the ubiquitous article "al-" occurring before a very large number of nouns in any text. It should be everywhere, unmistakable and impossible to miss. However, in the Voynich ms text, the VCI representation of <al>/<l>, which is the equivalent of EVA [ar]/[or], occurs predominantly at the ends of vords, not at the beginnings of them. A hypothesis of a "right-to-left" reading of the Voynich ms text is implausible, and in particular all the Arabic words cited above are the result of left-to-right reading of the VCI transcription. The only explanation I can think of to justify the Arabic hypothesis with the VCI character readings is that the author enciphered the text by writing the article "al-" at the ends of words rather than at the beginnings of words as in standard Arabic. This sounds bizarre, but after all we are dealing with a bizarre manuscript and text. One might perhaps compare it to the French argot "Verlan", in which the syllables of words are transposed. (Such an idea can also be considered for the many Romance languages with articles with "l", of which the word "Verlan" itself is one example.)
|
|
|
| Agreeing on standard transliteration files to use |
|
Posted by: Koen G - 16-09-2020, 04:42 PM - Forum: Voynich Talk
- Replies (23)
|
 |
In another thread, Rene suggested that we might better use standard transliteration files. This sounds like a good idea to me.
However, I have no idea about the variation in quality between the various options. I am just using Takahashi out of habit, but will gladly switch if this appears too flawed.
What are peoples' thoughts on this?
|
|
|
|
| [split] Verbose cipher? |
|
Posted by: geoffreycaveney - 15-09-2020, 10:05 PM - Forum: Analysis of the text
- Replies (123)
|
|
(15-09-2020, 09:22 PM)geoffreycaveney Wrote: You are not allowed to view links. Register or Login to view. (14-09-2020, 03:32 PM)RenegadeHealer Wrote: You are not allowed to view links. Register or Login to view. (05-08-2020, 09:57 PM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.I feel fairly strongly that "ain" should be analyzed without the "d" because there are many glyphs preceding the "ain" sequence. Get a grasp of "ain" first and then look at the letters in front, but not only the "d", the other ones too, so the pattern can be understood in context. I must say, it was pretty thrilling to read Koen's You are not allowed to view links. Register or Login to view. this summer about his experiment with ngrams, where one of the end results was some solid support for each in the series "a + 0~3i + line glyph" being one grapheme. I've been waiting for an experiment like that identifies ngrams with a high likelihood of being single graphemes ever since I read your old blog post about Janus pairs. I smell a breakthrough. I digress. Indeed, Koen's blog post that you link to above is fascinating. (It should be noted that Marco contributed significantly to the results there as well.) Has that particular topic been discussed in its own thread on this forum?
Summary of the main result of Koen's blog post for those who don't have the time or inclination to wade through it:
If one selectively considers each of the following character groups as a single character or letter, one transforms the ms text such that both its "h1" character entropy and its "h2" conditional character entropy (one critical problem with the Voynich ms text) are much more in line with typical h1 and h2 values of many natural language texts, in particular for European languages:
[ch], [sh]
[ain], [aiin], [aiiin]
[air], [ar], [al], [am]
[or], [ol]
[ok], [ot], [od]
[qo], [qok], [qot]
These substitutions -- treating each of the above character groups as a single character or letter -- generate an "h2" conditional character entropy of 3.01 and an "h1" character entropy of 4.12, which are actually within the normal range for many European natural language texts.
The most surprising result of such an analysis of the Voynich ms script would be that EVA "[a]" and "[o]" would virtually always occur as part of the above bigrams/trigrams -- they account for virtually all occurrences of [a] and [o]. They would thus have no more independent significance than do "[c]", "[h]", "[i]", or "[n]". This is surprising upon first consideration.
EVA [l] and [r] would still have a restricted independent significance primarily where they occur in vord-initial position, without a preceding [o] or [a]. I calculate approximately 1,300-1,350 such examples of [l] and approx. 400-450 such examples of [r].
I also suppose that according to Koen's schema, [aiir], [oir], and [oiir] should also be treated as single characters or letters. Probably they were simply too rare in the ms text to affect the h1 and h2 values very much, so they weren't worth considering in Koen's initial calculations.
Taking account of the entire resulting inventory of the remaining single characters, plus all of Koen's bigrams, trigrams, and n-grams above, in this way we have produced a character "alphabet" of about 25 values:
[e], [y]
[d], [k], [t]
[s], [l], [r]
[ch], [sh]
[ain], [aiin], [aiiin]
[air], [ar], [al], [am]
[or], [ol]
[ok], [ot], [od]
[qo], [qok], [qot]
Of course there are a handful of additional rare characters/letters as well. But the above inventory should account for the overwhelmingly vast majority of the Voynich ms text.
|
|
|
|
| Pisces 70v |
|
Posted by: Monica Yokubinas - 15-09-2020, 08:05 PM - Forum: Astrology & Astronomy
- No Replies
|
|
Pisces is the only one of the Zodiac pages that appears to have a proper name, 'Achazuz'
The word for Pisces/fish in Hebrew is Dg, like Dagon the fish God.
Back to the Voynich the first part of 'achaz' which means seized or caught. similar to the Arabic name of Al Risha for cord or rope that binds the fishes.
The zuz part of the word represents a old word in Hebrew for a very small coin. so for me this didn't make sense.
However I did find some interesting results from other languages that use a similar word for fish or Pisces:
You are not allowed to view links. Register or Login to view.
Lithuanian Zuvis
Latvian Zivis
A bit of history: Lithuania was known as the last of the pagans, since they were not Christianized until 1386.
Jerome of Prague went on multiple excursions to Lithuania, where he would burn down sacred tree groves.
Eventually he was kicked out of the country and later burned at the stake. An interesting coincidence indeed.
|
|
|
|
| previous Basque theories of the Voynich ms? |
|
Posted by: geoffreycaveney - 13-09-2020, 01:28 AM - Forum: Analysis of the text
- Replies (31)
|
|
As members of this forum know, last year I spent a considerable amount of effort arguing for a Judaeo-Greek interpretation of the Voynich ms, according to which a Greek-speaking Jewish person wrote the text in a similar manner as Greek Jews wrote their language using the Hebrew alphabet. I argued that the unusual features and patterns of characters in the ms could possibly be connected to the similarly unusual manner in which the Greek language was thus expressed with the Hebrew script. The fundamental problem with the hypothesis was the potential ambiguity of the phonetic values of many characters, creating ambiguity in the possible readings of many words. It was this obstacle that made it difficult to convince most people on this forum of the validity of my theory.
I always come back to the same problem: When one looks at the restrictions on character patterns, the small number of truly distinct characters occurring with any significant frequency in the ms text, and the well-known extremely low entropy of the text (the character patterns and sequences are too predictable), it is difficult to identify a language that could satisfy all of these restrictions in its writing system. I believe it was Rene who identified Hawaiian, with its extremely small and simple letter and sound inventory, as the natural language with entropy statistics closest to those of the Voynich ms. Very few languages spoken and known in Europe in the late medieval period would have resembled Hawaiian in this way.
Recently I stumbled quite by accident (in the course of researching a completely separate linguistic project) upon a description of the sound system of ancient "Pre-Basque", from which modern Basque dialects are descended. (Please note: I feel it necessary to state upfront and for the record here that I do NOT believe Basque is related to any other known language. It is a true language isolate. My other project was NOT about relating Basque to any other language or family.) I was struck by the simplicity of the Pre-Basque sound system:
Only 6 consonants may occur in word-initial position: *b, *g, *z, *s, *l, *n
*m did not exist, and *p was marginal
*t, *k, *tz, *ts contrasted with *d, *g, *z, *s only in intervocalic medial position
*r occurred only in medial and final position
The vowel system was a simple *a, *e, *i, *o, *u system
Well this is not quite Hawaiian, but it is a lot simpler than most European or Middle Eastern languages spoken and known in late medieval Europe. I would argue it is even simpler than the system I analyzed and proposed for written Judaeo-Greek. One could express the sounds and words more or less distinctively with a rather small and restricted character set, perhaps as small and restricted as that of the Voynich ms.
I should note that one Basque scholar I asked has already expressed skepticism, noting that by the late medieval period already Basque dialects had a consonant inventory as large as 18-20 phonemes, quite different from that of Pre-Basque described above. He found a provisional trial rendering of a few lines of Voynich ms text as Basque not to be intelligible. So this skepticism of one scholar should be duly noted.
However, I observe that the first published Basque text, Bernard Etxepare's Linguae Vasconum Primitiae (1545), in its original spelling used the Latin letter "c" to express both Basque "k" and some instances of modern Basque "z" (just as English uses "c" for the sounds "k" and "s"), and it only needed to use 10 consonant letters with any significant frequency:
r, c, n, d, g, t, s, b, l, z
(m, p less frequently, f, q very rarely)
Moreover, strikingly, in one excerpt of this text in its original spelling that I examined, "n" appears almost exclusively either in word-final position or in the coda position of a syllable, before a consonant that begins the following syllable. This would be a striking correspondence with the restricted occurrence of the Voynich ms character also transcribed as [n] in the EVA transcription.
I am curious to know about any previous theories of the Voynich ms as a text written in Basque. I would have expected this to be a relatively popular hypothesis as far as these things go: Many researchers (read: many well-respected analysts including some on this forum who disagree strenuously with any Latin, Romance, or Greek theory including mine) argue that the linguistic structure of the Voynich ms text is unlikely to represent an Indo-European language; it is fairly clear that the ms is of European provenance; and Basque is one of the very few non-Indo-European languages that was spoken indigenously in Europe in the late medieval period.
However, Internet searches turn up surprisingly few references to any previous Basque theories of the Voynich ms as far as I can tell. It is true that we apparently have no extant manuscript length written Basque texts before the mid-16th century publication mentioned above. But some Basque was written in the late medieval period, so we know there were literate Basque speakers. In a certain way the lack of a standard literary form of written Basque in this period could be an argument in favor of the theory: It would have given the author of the ms more license to develop the writing system in the ms more freely in an idiosyncratic style, without constraint or influence of a literary standard. Surely there were Basque speakers in Europe in that period who were literate in Latin, and who would have had the ability to compose such a ms in Basque.
I suspect the main reason there are few Basque theories of the Voynich ms is that few non-Basques can speak or read Basque! It seems to me that even among Europeans with an interest in unusual or less commonly studied languages, the idea of studying Basque may be more popular than the actual studying of Basque. It is naturally quite difficult to learn after all, and it cannot be related to any other known languages, so there is no natural linguistic link or gateway into studying Basque. It is a linguistic field of study sui generis. Few people would be well qualified to investigate, analyze, and evaluate such a theory.
For the record, here is a very rough example of how one may formulate a very provisional proposal to represent many of the most frequent and distinct sounds of Basque with the Voynich ms characters. I caution that at this stage such a proposal is very unlikely to be completely correct, and I further caution that as noted above, one Basque scholar has already rejected a possible interpretation of a few lines of the ms according to this schema. But it is a starting point and an example of one possible such system.
EVA [a] = Basque "a"
EVA [ai] = Basque "e"
EVA [e] = Basque "i"
EVA [o] = Basque "u"/"o"
EVA [n] = Basque "n"
EVA [k]/[f] = Basque "d"/"t" (always "d" as initial)
EVA [t]/[p] = Basque "b" (Basque "p" is rare/marginal in any case)
EVA [d] = Basque "c" in Etxepare 1545 (either "z" or "k"/"g" in modern Basque - see above)
EVA [ch] = Basque "h" (This phoneme is not even considered for alphabetization in Basque etymological dictionaries)
EVA [ckh]/[cfh] = Basque "t"/"d" (in principle "t", [ch] representing aspiration, but likely not so clear in the actual ms)
EVA [cth]/[cph] = Basque "p"/"b" (in principle "p", [ch] representing aspiration, but likely not so clear in the actual ms)
EVA [s] = Basque "z"
EVA [sh] = Basque "s"/"z" (in principle "s", [ch] representing voiceless articulation, but likely not so clear in the actual ms)
EVA [l] = Basque "l"
EVA [r] = Basque "r"
EVA [y] = abbreviation for some word endings and prefixes, in the style of the similar-looking medieval Latin ms abbreviation
As any experienced Voynich ms researcher might guess, this schema will work out much more messily in practice as a reading of the actual ms, than it appears in this more or less neat correspondence table. I had to shoehorn Basque "g" into EVA [d] along with the "k" and "z" that Etxepare represented with "c" in 1545. [y] is a big fudge factor in the above schema, as it can represent an abbreviation for any ending or prefix as it stands now. Finally, to make any sense of many words, it would be necessary to add a rule that many final vowels are simply unwritten in this system. This is possible in rudimentary writing systems, but it is another fudge factor.
Of course I have my reasons for making the educated guesses about the letter/sound values of the characters above, but obviously everything must be considered provisional at this stage (understatement, I know).
I have chosen EVA [d] to represent Basque "c" and thus often "z" because among the very most frequent Basque words are the 3rd person past tense 'to be'/'to have'/auxiliary verb forms "zen" ('was'), "ziren" ('were'), "zuen" ('had'), and "zituen" ('they had'), and the participial form "izan" ('being', 'having') is very frequent as well. It should be noted that such an auxiliary verb form must be used in almost any Basque clause with a finite verb -- most lexical verbs themselves do not conjugate as Indo-European verbs do.
If EVA [d] is "z" and EVA [n] is "n", then this means that the most frequent Voynich ms word [daiin] could possibly represent the Basque word "zen" ('was'), the 3rd most frequent word in Basque corpora. ("zuen" is 4th, and "izan" is 5th.)
The most frequent Basque word "eta" ('and') could possibly be accounted for by the well-known interpretation of the Voynich ms character EVA [q] as the equivalent of an ampersand.
It is more difficult to explain the 2nd most frequent Basque word "da" ('is'), the 3rd person singular present tense 'to be'/auxiliary verb form. Perhaps this might possibly imply that the Voynich ms is primarily written in the past tense rather than in the present tense. This may strike us as rather implausible at first consideration, at least for the herbal, pharmaceutical, and recipes sections. The verb forms in the recipes section, however, may be primarily imperative forms rather than indicative present tense forms.
I have often thought that the repetitive and line-based patterns of much of the Voynich ms text may be more plausibly interpreted as poetry rather than prose. (I note for the record that Etxepare's Basque publication in 1545 was a collection of poems.) If the "herbal" section is actually poetry inspired by plants, rather than an encyclopedic type of herbal manual, then perhaps a large part of this poetry is written in the past tense. Alternatively, perhaps the Basque participial form "izan" ('being', 'having') features more prominently in the herbal and pharmaceutical sections than do present tense forms.
The other very frequent Voynich ms word [chedy], found most frequently in the balneological and recipes sections, might possibly represent the Basque word "hik", the ergative form of the familiar 2nd person singular pronoun 'you'. It would be logical if this word were frequent in these sections -- in dialogue in the balneological section, and in instructions in the recipes section -- but much less frequent in the herbal section. Common variations of this word could possibly represent similar Basque words such as "hi" (absolutive 2.sg. 'you'), "heuk" and "heu" (emphatic 2.sg. ergative and absolutive 'you'). Likewise [shedy] and related words could possibly represent the formal, neuter, and plural variations of this pronoun in Basque that begin with "z" rather than "h".
Let me conclude by repeating and emphasizing the tentative nature of this hypothesis and all of these ideas and comments. I have been wrong about such things before, and I am well aware that I may well be wrong again now. But I find the theory intriguing, and I welcome comments and feedback from members of this forum. Naturally, if we have any Basque speakers on this forum, their feedback and reaction to my ideas would be the most valuable of all.
Geoffrey
|
|
|
|
| What evidence would it take for you to change your mind? |
|
Posted by: Mark Knowles - 07-09-2020, 06:07 PM - Forum: Voynich Talk
- Replies (9)
|
|
"What evidence would it take for you to change your mind?" this the classic question that one should pose someone or oneself with a theory and given that there are many Voynich theories a question that one should be ready to pose to a theorist.
Now another question is how one can expect someone to answer that in such a way that it allows the realistic possibility of disproving their theory.
In terms of some of my own ideas, which it must be said do not include a decipherment, I have been finding it hard to provide an answer which allows scope for reasonable attempts to disprove them, which is frustrating.
Clearly if the manuscript was correctly deciphered it would provide much scope for disproving various theories, however that is a big ask. It seems all the different evidential options for disproving my ideas seem to not be easily attainable. It seems the options for proving are on paper a bit more attainable, but not necessarily easy to obtain at all.
How do we think a proposed decipherer can honestly answer that question in a way that it provides scope for disproving their theories?
|
|
|
|
| The Linguistics of the Voynich Manuscript (Bowern et al. 2020) |
|
Posted by: Stephen Carlson - 07-09-2020, 06:52 AM - Forum: News
- Replies (45)
|
|
There's a new paper on the VM at You are not allowed to view links. Register or Login to view.:
Quote:You are not allowed to view links. Register or Login to view.
You are not allowed to view links. Register or Login to view., You are not allowed to view links. Register or Login to view.
September 2020
The Voynich Manuscript is a 15th Century illustrated cipher manuscript. In this overview of recent approaches to the Voynich manuscript, we summarize and evaluate current work on the language that underlies this document. We provide arguments for treating the document as natural language (rather than a medieval hoax) and show how we can make statistical arguments about the phonology, morphology, and structure of the document, even though the contents remain undecipherable.
ETA: Claire Bowern is a legit linguist, so I would expect this paper to be a serious effort.
|
|
|
|
|
): Adyghe, Ag(h)ul; Andi -- related; Ossetic -- not as much