The Voynich Ninja

Geoffrey,

We are all searchers and we often change our minds after new discoveries.
Judaeo-greek or judaeo-spanish is not a problem for me.

I don't know if it's technically possible but it should be a good idea to change the title of this thread into :
geoffrecaveney's theory, as it was done for Morten St. George theory.

(26-03-2019, 11:44 PM)Paris Wrote: You are not allowed to view links. Register or Login to view.Geoffrey,

We are all searchers and we often change our minds after new discoveries.
Judaeo-greek or judaeo-spanish is not a problem for me.

I don't know if it's technically possible but it should be a good idea to change the title of this thread into :
geoffrecaveney's theory, as it was done for Morten St. George theory.

Paris, thank you for your open-mindedness, but for now, I believe the current thread title is still the best summary description of my theory. I still find only (quasi-)Judaeo-Greek text in my reading and interpretation of the Voynich ms text. The other text I posted and commented on was a separate item.

Geoffrey

Rene in the "Successful translation ?" thread stated that a proposed solution should include letter/character correspondence tables showing how the original encryption steps were done.

I described the details, reasons, phonological, orthographic, and historical explanations for most of the encryption process in a previous post in the current thread, at this link:

You are not allowed to view links. Register or Login to view.

Now I have created a Greek : Judaeo-Greek : Voynich letter correspondence table to present to readers of this forum.

The previous post linked above should be considered as essentially a set of footnotes to this table.

One very important detail not included in the previous post: the explanation of the Voynich character [ch]. 

First, Voynich [ch] represents the Judaeo-Greek Hebrew letter "he" ("h"). Greek has never had a letter for the "h" sound, but sometimes pronounced it before word-initial vowels. 

Second, [ch] must be considered as an alternate character for [e]. (Also, [i] is a much rarer alternate character for [e].) The author encrypted Greek iota or Judaeo-Greek yod as either of these characters. For example, in word-initial position, the author almost always used [ch] rather than [e]. (This is consistent with the Greek "h" sound before an initial vowel!) If you have studied some hard-to-read characters and words in the Voynich MS, you have probably found instances where it is hard to tell whether a character is [ch] or [ee] (that is, connected or unconnected). The relationship of [e] and [ch] proposed here is consistent with this orthographic ambiguity in the ms text.

Since Judaeo-Greek also sometimes wrote yod for Greek upsilon, eta, and epsilon, these letters may be encrypted as either [e] or [ch] as well.

Also, if you know that the popular Greek food "gyro" is pronounced "euro" or "yiro", then you know about the sound change of Greek "g" > "y". Likewise, in my reading of the Dionysian term "sparagmos" as the last word (right-justified) at the end of the first paragraph, for example, the Greek "g" is represented only by a Voynich [i]. This reflects a "y" pronunciation rather than the "g" Greek spelling.

One other minor change from the previous post: the interpretation of the Voynich character [f] is difficult because it is so rare, but the correspondence table shows that the encryption process is more consistent if Judaeo-Greek "b" rather than Judaeo-Greek "ph" is the letter sometimes represented by Voynich [f]. In any case, [p] is more common, and as I explained in detail in the previous post, in most places in the ms this whole series of labial consonants became Voynich [d], due to the identical pronunciation of the Judaeo-Greek letters "beth/veth" and "vav" as the sound "v".

=====

General comment on this correspondence table: Notice how the second step, from Judaeo-Greek to Voynich, is not actually any more convoluted than the first step from Greek to Judaeo-Greek (without the vowel diacritic dots written, that is). And the first step from Greek to Judaeo-Greek is a known and established historical linguistic fact, not just a part of my theory! This table shows that once one has made the step from Greek to Judaeo-Greek, it is not such a giant leap to make the additional step from Judaeo-Greek to Voynich. Yes, the consonants are further ambiguated. But plenty of letters were already ambiguated going from Greek to Judaeo-Greek!

I already thoroughly explained the reasons for the ambiguation of Greek vowels in Judaeo-Greek without diacritic dots. But the reader should also note certain ambiguation of Greek consonants in Judaeo-Greek as well:

Hebrew never had separate letters for the fricative sounds "kh" (/x/), "th", and "ph" (/f/). Historically, the Hebrew letter "chet" had represented a pharyngeal or uvular fricative, not the velar fricative /x/ that speakers of European languages are more familiar with. In Hebrew, these fricative sounds are just allophonic alternative pronunciations of "k", "t", and "p" in certain places, particularly in medial and sometimes in final position. (It is likely that Hebrew borrowed this alternation from Aramaic.) So in Hebrew, the voiceless stop and the fricative are written with the same letter.

But in Greek, the distinction between "k" and "kh" (usually written as "ch" in Latin letter borrowings), "t" and "th", and "p" and "ph" is phonemic: the distinction is important in distinguishing meaning between different roots and words. When Judaeo-Greek speakers wrote their Greek language in their Hebrew script, they had no natural way to make these distinctions with separate letters, since Hebrew lacked letters for the fricatives. Thus in Judaeo-Greek, as the table shows, you see the Hebrew letter "kaf" representing both Greek kappa and Greek chi; the Hebrew letter "tav" representing both Greek tau and Greek theta; and the Hebrew letter "pe" representing both Greek pi and Greek phi. Now the Judaeo-Greeks did also adapt the letter "qof" for kappa sometimes, and the letter "chet" for chi sometimes, and the letter "teth" for tau sometimes. But nevertheless, the ambiguity of "kaf", "tav", and "pe" remained, as the table reflects.

Thus, some of the ambiguity in consonants that I identify in the Voynich script, may have had its roots in this certain measure of ambiguity created by representing Greek with the Hebrew script in Judaeo-Greek. Naturally, it is clear that the Voynich ms author added more ambiguity.

So then, without further ado, I present to readers of this forum the Greek : Judaeo-Greek : Voynich letter correspondence table:

[attachment=2740]

-----
Geoffrey Caveney

Thanks very much, this is very constructive!

Years ago when I was trying to resolve the VMS text into a long list of languages one by one, I didn't know about Judeo-Greek (I tried Ladino, Yiddish, etc., but somehow, even though I was looking for different adaptations of Jewish languages, I missed Judeo-Greek).

Not knowing about Judeo-Greek, I was simply trying Hebrew, Ladino, Yiddish, Greek, Amharic, Latin, Estonian, Icelandic, Persian, etc., as separate languages.

So, trying for just Greek, I would get things like this (I had to dig through my files to find these, it's so long ago):

υαερτα γουαρίς υαρίς (or maybe υερίς) ωτορ οχς

I didn't treat the VMS text as an abjad even though it has always struck me as maybe having abjadic qualities, I was simply trying various substitution schemes. No extra characters were added, this is straight Voynichese (I'm not going to say which folio it's from, yet).


On the surface, it almost looks like Greek and can almost be read as Greek (I didn't even tweak this to try to make it work better), and it wasn't too hard to get phrases like this that were almost a line long, but when one looks at it more closely, it didn't strike me as legitimate Greek (with my very limited acquaintance of Greek).

Also, there were chunks in between these phrases that didn't seem to make any sense at all.

One important point to add:

For those who are familiar with Voynich word structure, it will quickly become clear from a brief overview of the table that with this correspondence system, very few Voynich words will end with the equivalent of a Greek vowel. At first glance, this seems implausible, since so many Greek words end in vowels. However, there are at least two logical explanations for this:

1) The obvious one, in Judaeo-Greek, is that the final vowel was indicated only with vowel diacritic dots beneath the final consonant, and in this text the vowel diacritic dots are not written.

2) The other possible explanation invokes the Ancient Greek phonological phenomenon known as "movable nu":

You are not allowed to view links. Register or Login to view.

In brief, in Ancient Greek there were a significant number of words where a letter nu, which was not inherently part of the word, could be added to the end of the word, for prosodic purposes. In all of the examples in the article, this movable nu follows a final iota or a final epsilon.

It is possible that the Voynich ms author made extremely liberal use of this movable nu, adding it to virtually every Greek word ending in iota or epsilon, and perhaps eta or upsilon as well, or perhaps any vowel. This would explain the very large number of Voynich words that end in one of the two characters [l] and [(i)in].

It is already clear from my readings and interpretations so far that, if my theory is correct, the author was quite familiar with many Ancient Greek forms. My interpretations so far produce a mixture of classical and medieval Greek forms, words, grammar, and syntax. Again, this is not necessarily atypical of Byzantine Greek, although I am not claiming that the particular style of mixture of forms that I find here is necessarily akin to any known extant style of Byzantine Greek.

One may fairly ask, how likely was it that an early 15th century Judaeo-Greek speaker, whose literary training and experience would have been more likely to be in Hebrew, would also be familiar with some of the finer points of Ancient Greek forms, grammar, style, and syntax?

Thus I should refine the hypothesis in this way: The Voynich ms author was *familiar* with Judaeo-Greek and the way that it was written in the Hebrew script. The author could have been a Judaeo-Greek speaker himself or herself, or the author could have simply been any Greek speaker who became familiar with Judaeo-Greek and its script. The ambiguity of Greek vowels and some Greek consonants in the Hebrew script of Judaeo-Greek could have been an inspiration or a motivation or a basis for this author to develop the Voynich script. The author may have been more knowledgeable about classical and medieval (modern to the author) Greek, and then simply used the Judaeo-Greek script as a basis from which to create the Voynich script.

Geoffrey

P.S.: It will get even more complicated if the author added the concept of a "movable sigma" to add an extraneous sigma (Voynich [y]) to the ends of some Greek words that don't inherently have the letter. In fact, in the phrase of Judaeo-Spanish that I interpreted from the other text that I posted here, I noticed the unusual endings of the last two words of the phrase "todos dias de-tus vidas". In standard Spanish, this last part "tus vidas" sounds very strange to me: It combines singular "you" and plural "lives"! In Judaeo-Spanish, one explanation is that final "s" had a tendency to be dropped in pronunciation, a very familiar phenomenon in Andalusian, Latin American, and above all Caribbean Spanish. If final "s" is essentially silent anyway, adding an extraneous final "s" in writing doesn't affect the pronunciation at all! In any case, this possible example of extraneous final "s" in Judaeo-Spanish may offer a possible explanation for a parallel phenomenon in the writing of Greek words in the Judaeo-Greek or Judaeo-Greek-inspired script of the Voynich ms.

Reading, interpretation, translation, and analysis of the first line of the very last paragraph of full text in the Voynich MS, folio page You are not allowed to view links. Register or Login to view. (the last paragraph is only the final five lines, as the preceding line is short):

text:
[ oqokain  al  sh2ey  qokar  okaral  okey  sh1cphhy  oteey  o  okar  otydy ]

Note: In my own transcription I use [sh1] to indicate [sh] with a more open (really just less closed) loop on top, and [sh2] to indicate [sh] with a more closed loop on top.

reading:
" ok-odAn  an  meis  kOthar  OthAlAm  Otes  ti-pies  Okheis  o  OthAl  oksus "

Note: This line suggests to me that Voynich [q] is just an alternate character for Voynich [t] (Greek kappa/chi/gamma). This is consistent with the rest of the correspondence system, as in my table Voynich [k], [p], and [l] already had alternate characters, but [t] did not. Now it is more in parallel with the rest of the system.
"A" represents a letter that could be represented by the Hebrew letter aleph in the Judaeo-Greek script; "O" represents a letter that could be represented by the Hebrew letter ayin in the Judaeo-Greek script. There is no phonemic or phonological distinction between these two letters in Judaeo-Greek. The difference is essentially etymological in Hebrew, but this has no relevance to Greek words written in Judaeo-Greek. They are simply silent "placeholder" letters in the Judaeo-Greek script, beneath which a variety of vowel diacritic dots can be written. But in the Voynich MS text the vowel diacritic dots are not written.

interpretation:
" ouk-ouden  an  [e]meis  kathar'[a]  ethelom[en]  autes  ti'-[o]poies  ekheis  ou  ethel'[ei]  oks[e]os "

" ουκ-ουδεν  αν  [η]μεις  καθαρ'[α]  εθελομ[εν]  αυτες  τ[font=Arial, Helvetica, sans-serif]ι'-[ο]ποιες  εχεις  ου  εθελ'[ει]  οξ[ε]ως[/font] "

Note: I will discuss the syntax of the Greek sentence below. I note here that the two omitted final vowels at the ends of καθαρ'[α] and εθελ'[ει] can be seen naturally in Greek as elision before a following initial vowel.
One technicality about the letter correspondence table: I neglected to indicate that the Greek diphthongs "ου" and "αυ" can be represented simply by [o] in the Voynich script. This strikes me as rather natural, more so in fact than some of the other ambiguities in the representation of Greek vowels created by the encryption step from Greek into Judaeo-Greek without vowel diacritic dots.

literal translation:
" not-at-all (particle) we purely [don't] want those whom you-have not wanted keenly "

freer translation:
" we purely and absolutely do not want those whom you have not wanted keenly "

Analysis:

To me this line reads as the author addressing a deity: Those whom the deity does not want, we the devotees of the deity do not want them either.

It is also possible to interpret the meaning as "We do not *wish for those things* which you have not wished for". But the Greek forms for "those whom" / "those which", "αυτες  τ[font=Arial, Helvetica, sans-serif]ι'-[ο]ποιες[/font]", have distinctly feminine plural suffixes, which strikes me as non-standard as an expression of "those things", for which I would expect neuter suffixes.

Regarding the Greek syntax: Again it is a mixture of classical and medieval/modern forms, but this sentence seems to me to flow more smoothly, with less of the information in the grammatical suffixes missing.

The intensive double negation (negative concord) at the beginning of the sentence uses classical Greek negative forms, and the negative particle near the end of the sentence is also a classical form. I note that the intensity of the negative expression at the beginning is matched by the use of the rare Voynich word-initial [oqo-] as opposed to the vastly more common [qo-].

I interpret both instances of the word [okaral] / [okar] in this sentence as the classical Greek verb form εθελ- "want".
The last two letters of εθελομ[εν] ("we want/wish") appear to be omitted, but the Voynich character [l] represents both Greek mu and Greek nu, so arguably here it represents both the "μ" and the "ν" in this suffix. Further, the letter "μ" in this suffix alone distinguishes this form as 1st person plural. The only possible confusion could be with the 1st person singular form of the middle voice, but the explicit overt expression of the subject "we" with the word [η]μεις makes such confusion very unlikely to the Greek reader.

Most of the rest of the sentence seems to have more modern syntax: 

καθαρ'[α] : adverb, "purely, neatly, plainly, clearly"
Strikingly, the adjectival form of the same word is also the Greek translation of "kosher" !

αυτες reflects modern, not ancient, pronunciation of this form

[ο]ποιες is the modern, not ancient, Greek relative pronoun. The spelling diphthong "οι" is one of the many Greek vowels that underwent iotacism and is simply pronounced like iota, so the spelling in this MS reflects that correctly. The uncommon extension of Voynich [-cph-] to [-cphh-] reflects the consecutive syllables with front vowels in [ο]ποιες.

The prefixed τ[font=Arial, Helvetica, sans-serif]ι'-[/font] before -[ο]ποιες is tricky. Typically, yes, this Greek relative pronoun does standardly take the definite article before it, so the τ[font=Arial, Helvetica, sans-serif]ι'-[/font] is expected and correct in this sense. (The full form for the feminine accusative plural would be τ[font=Arial, Helvetica, sans-serif]ις[/font].) But the 3rd person pronoun αυτες is the immediate antecedent, and it is almost identical to the definite pronoun, with a prefix attached. Rather than repeat the full syllable with αυτες [font=Arial, Helvetica, sans-serif]τ[font=Arial, Helvetica, sans-serif]ις [font=Arial, Helvetica, sans-serif][ο]ποιες[/font][/font][/font], the author simply abbreviated the definite pronoun to τ[font=Arial, Helvetica, sans-serif]ι'-[/font] as a prefix.

The analytic or periphrastic compound verb construction εχεις ου εθελ'[ει] is a curious combination of the ancient Greek negative particle ου and the more modern periphrastic formation of the present perfect tense by εχεις [font=Arial, Helvetica, sans-serif]εθελ'[ει][/font] ("you have wanted/wished").

-----
Geoffrey Caveney

Here is a revised and updated version of the letter correspondence table, accounting for the addition and clarification that I explained in my notes to the reading and interpretation of the first line of the last paragraph of the Voynich MS on folio page You are not allowed to view links. Register or Login to view. .

It still remains to place Voynich [m] and [g] in particular in the correspondence table. So far in passages on other pages my best reading and interpretation has been that they are likely to correspond to Greek sigma, thus representing alternate characters for [y]. But I do not want to rush to this conclusion. Voynich [g] is very rare, occurring less than 100 times in the entire almost 38,000-word MS. And Voynich [m] is well-known to be primarily a line-ending character. As such it may tend to represent line-final word endings in a more general way.

Also, I am curious about the lack of an alternate character for Voynich [r] (Greek lambda and rho), since all the other consonant series characters now seem to have them (if [g] and/or [m] are alternates for [y]). But I have not seen evidence for the existence of such a character yet. Perhaps it is possible that yet another form of the loop on top of the [sh] character could possibly prove to be such a character. But trying to clearly determine the distinction between two different forms of open (less closed) loops on top of [sh] strikes me as quite difficult. So far all instances of [sh] with open loop on top have made more sense as Greek tau/theta/delta.

[attachment=2741]

-----
Geoffrey Caveney

Benched characters? I can't remember if you covered them in your previous posts.

(28-03-2019, 05:01 PM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.Benched characters? I can't remember if you covered them in your previous posts.

I take it you are referring to the Voynich characters which in EVA transcription are written as [ckh], [cth], [cph], [cfh], the apparent ligatures of [k, t, p, f] with [ch].

At this stage, I am interpreting them simply as [k]+[ch], [t]+[ch], [p]+[ch], [f]+[ch].

Now it would be plausible that these characters would be more likely to represent the Greek fricatives theta, chi, and phi. That would be a logical use of [ch] as the "h" character.

However, I also find plenty of examples of Greek theta and chi represented simply as [k] and [t] respectively. For example, in the first line of the last paragraph of You are not allowed to view links. Register or Login to view. , [qokar] = καθαρ', [okar-] = εθελ-, and [oteey] = εχεις.

Moreover, the one example of one of these ligatures in this line, in [sh1cphhy], is not a Greek fricative but rather Greek pi in τ[font=Arial, Helvetica, sans-serif]ι[/font]'-[ο]ποιες. Here the extended [c_hh] part of the ligature simply represents the consecutive syllabic front vowels iota and epsilon in this Greek word. (Recall that as a result of the iotacism sound change, Greek "οι" is simply pronounced /i/ like iota.)

At the same time, in this same line [k] also represents Greek delta in [oqokain] = ουκ-ουδεν, and Greek tau in [okey] = αυτες. The characters are ambiguous; they may represent the voiceless stop, the voiced stop, or the fricative. That is an inherent feature of the system, for better or worse. We have to deal with it. 

[Note: the Greek "voiced stop" letters beta, gamma, and delta are actually pronounced as voiced fricatives, or in the case of gamma also as a "y" glide, in medieval and modern Greek.]

Geoffrey