Voynich manuscript decoded?
- Thread starter 332nd
- Start date
shadron
Philosopher
- Joined
- Sep 2, 2005
- Messages
- 5,918
That gives him phonetic values for some of the characters.
Aepervius
Non credunt, semper verificare
Which is not that much different from, all previous attempt, with schemes giving *some* recognizable name but never allow a full text understanding, or even more words or paragraph meaning.
If some of the characters have been deciphered then some elucidation of the manuscript must have been achieved. With "some" characters one gets a few words. That happened in the 1920s with the decipherment of Ugaritic, a Bronze Age semitic language written in a cuneiform alphabet. See http://images.library.wisc.edu/WI/EFacs/transactions/WT1966/reference/wi.wt1966.adcorre.pdf
Aepervius
Non credunt, semper verificare
That applies only if you really are deciphering recurrent word and structure. The problem with the titular manuscript of the thread is that the claim of deciphering is usually in isolated word, which are taken as being similar to word in other language referring to plant or whatnot. The deciphering is never about structure of the sentence or "regular" words.
And that reeks of texas sharpshooter fallacy , finding something similar to what you know, claiming victory, but it does not advance the real deciphering a single bit.
Monketi Ghost
Confusion Reactor
- Joined
- May 21, 2003
- Messages
- 25,141
I am inclined to think that given the number of serious efforts over the years, if there's any sense to be made of it, that would have happened by now.
It's a for-the-ages troll-job.
It's a for-the-ages troll-job.
Delvo
Дэлво Δε&#
Talking about what "usually" happens does not make a valid comment on what actually is happening in this particular case. Bax's process isn't like your description of the "usual" at all. It involves comparing multiple candidate plant names for sounds & symbols they have in common, which can then be applied to any other word in the book that has those symbols, which partially fills those other words in and provides hints to the remaining letters, which also can be applied to other words, and so on, so the scope keeps widening. The lists of letters and plant names that seem to have been worked out so far has been growing since this paper was first published, and now includes some astrological entities and a few words that aren't plant names but are related: two words for plant parts, one for dried plant matter in general, and one for a category of multiple plant species. The inherent cross-checking and back-checking subjects every idea along the way to possibly being contradicted by new information.
Well, of course not. That kind of stuff is at the opposite end from the starting point. That's the stuff you work your way up to gradually, after starting with nearly nothing and passing through intermediate levels on the way first. When you're still working on the alphabet, how in the world could you possibly already know anything about sentence structure, or about words you can't tie to an external information source for meaning? Where is this knowledge supposed to come from? To even imagine expecting anything like that is to have just not paid any attention to what the process actually is.
This method hasn't been tried before.
novaphile
Quester of Doglets
It looks like a herbalist's handbook...
Since other people haven't commented on this part, I thought I would chime in.
The small part I've had a look at, on Stephen Bax's website, sits very well with the idea that it is a herbalist's handbook.
The pages on plants seem to start with multiple names for the same plant, which Professor Bax identifies as being common practice for other mediaeval plant texts...
So, after very little reading... I'd guess that the information about stars, constellations etc. would be to do with the timing (or choice) of treatments or when plants should be gathered for the purpose of making medicines (potions or poisons).
So far it doesn't look like a code, as such, more likely just an unusual language.
Some of the commentary suggests that some of the plant illustrations may have been copied from other illustrations. The section of Dr Bax's website written by the unnamed Finnish Biologist is quite interesting in its own right.
Given the bad press associated with "Voynitch" I can understand why contributers would wish to remain anonymous.
Since other people haven't commented on this part, I thought I would chime in.
The small part I've had a look at, on Stephen Bax's website, sits very well with the idea that it is a herbalist's handbook.
The pages on plants seem to start with multiple names for the same plant, which Professor Bax identifies as being common practice for other mediaeval plant texts...
So, after very little reading... I'd guess that the information about stars, constellations etc. would be to do with the timing (or choice) of treatments or when plants should be gathered for the purpose of making medicines (potions or poisons).
So far it doesn't look like a code, as such, more likely just an unusual language.
Some of the commentary suggests that some of the plant illustrations may have been copied from other illustrations. The section of Dr Bax's website written by the unnamed Finnish Biologist is quite interesting in its own right.
Given the bad press associated with "Voynitch" I can understand why contributers would wish to remain anonymous.
I think the process you describe is very usual. In the case of Ugaritic an axe or adze was found with symbols inscribed on it. On the assumption that the text represented a Semitic word for axe, various letters were assigned phonetic values and it was soon found that applying them to other sets of symbols produced plausible semitic words, and the scope widened very quickly, so that effective decipherment was achieved in a few weeks or months. Now, if the assumption that Ugaritic was semitic had been wrong, no further progress would have been made, even if the meaning of word on the axe had been correctly divined.
Which of these scenarios applies in the Voynich case? It is not clear to me. Have any phonetic values been derived from assumptions about the plant names? If not, how can the plant names be said to have been deciphered? If yes, what are they, and what do they tell us about the character of the "Voynich language"?
The use of plant names in this context has a precedent in the Linear B decipherment. A word recovered very early in the process was ko-ri-ja-da-na, meaning coriander.
Last edited:
Delvo
Дэлво Δε&#
In serious linguistic work, yes, but I was answering someone talking about what's apparently usually been done with the Voynich manuscript. It apparently has attracted some cooks & quacks before, which causes some JREFF people to equate any serious study of it with the cooks & quacks, as if the fact that those exist made it impossible for anyone who isn't one to work on the manuscript.
Apparently both are the case, with different people's approaches. Bax's method is more like what you're talking about with Ugaritic, but his website also has some posts on it by other people who weren't using his method. The difference is self-illustrating; the latter don't go anywhere because there's no next step anybody can take with those starting points, while Bax's method continues to gradually yield more information and put its implications to the test.
Yes. Bax's original paper and video, linked at the beginning of the thread, showed how he came up with the letters for t, k, n, r, w, a, o, another vowel or two of uncertain nature, and x (a common phonetic symbol in linguistics for the fricative spelled "ch" in German and "kh" in transliterated Arabic). Carrying on since then, there's now also pretty good evidence in multiple words apiece that certain other letters were h, two sibilants, p, b, and g. There are also new one-shot hints at possible v and j and even some possible digraphs (sequences of two letters for one sound, like our "th" and "ng").
This is a good spot to point out an example of how Bax's method & results differ from what some here have misrepresented it as, apparently based on the fact that the scary magic word "Voynich" is involved instead of actually reading the paper or watching the video, nevermind following the progress since then at his website. There's one letter, the tall skinny one with two straight legs and two little loops on top and nothing overlapping with the bottom, that has three open theories about its sound right now.
Bax's original thought about it was that it might be "L" because of how often it shows up as the second letter (after "a") on the astrological pages, given the fact that Arabic star names usually start with "al". Someone else, after an argument that the manuscript showed a strong influence from Mandaean culture, pointed out that Medieval astrological writings often speak of gods or spirits associated with the stars & planets or with their associated times of the year, and that in Mandaic, the prefix for references to gods & spirits is "ab", so the letter was likely to be "b". But in the herbal section, we now have seven plant names/words so far where "g" is the best fit for that letter, along with nine where a different letter seems to equate to "b", and another issue with "L" that I'll bring up again below. I figure those multiple data points outweigh what seems to be just one prefix recurring a bunch of times in the astrological section. But the bottom line with Bax's method and the people who are joining in and following it is that more evidence will support the right one and contradict the wrong ones (and already has in another couple of other cases).
Disjointed bits & pieces so far. By far most of the apparent cognates are either Semitic or Indo-European. Those are big language families, so that doesn't narrow it down much, especially since the words/names we have so far are the most likely to have been imported from other languages instead of pure Voynichese anyway (which is the whole reason why we can find the cognates at all).
I subscribe to the theory that the alphabet is derived from the Syriac alphabet, based on graphical similarities between Voynich letters and their Syriac counterparts. If that's right, the changes in the alphabet are as drastic as what's happened to other alphabets when they jumped from one language family to another, so the language wouldn't be Semitic.
The ends of words are significantly more redundant than the beginnings, making it appear that the language uses suffixes a lot but prefixes little or none. I suppose it might be possible to narrow down the range of possible languages based on what those suffixes' sounds are, but it hasn't been done yet. Maybe it will when the this gets the right linguist's attention.
Unless the beginning of those words in the astrological section really is "al" and the apparent cases of the same letter as "g" in the herbal section are wrong, there doesn't seem to be a letter for the "L" sound. In a couple of names where it would be expected, there's nothing there, and in one or two more, the Voynich equivalent of "tw" is there instead. That's exactly what it would look like if Voynichese didn't even have the sound and was either dropping it or converting it to something else when it imported L-words from other languages.
That's one of the first round of plants Bax named in this case, enabled by the fact that it has sounds in common with the names of other identifiable Voynich plants such as kentaury and hellebore (which is called various names in the general form k_r or kh_r in a bunch of central & southern/southwestern Asian languages).
Using plants makes it more difficult than when they used people & places in previous cases like this, for three reasons. First, it helps to know what thing might be getting named, and identifying plants is hard to do even with real plants, even harder with rather weird drawings of them; most plants in the book have had multiple identifications suggested. Second, plant names in foreign languages are often more obscure and harder to find than names of places or famous people. And third, plant names that get transferred from one language to another can have already done so ages before the book was written and evolved a lot since then, whereas references to people and countries/cities are more likely to be "current events" of their times and thus less altered by time before getting written. Fortunately, one thing we have going for us in this case is that this book is big, with lots of data points to work with.
Nay_Sayer
I say nay!
I took all the information and decoded the manual:
It says "Drink more Ovaltine".
It says "Drink more Ovaltine".
Thanks for your interesting post. I have to say I'm not yet convinced that Bax has got anywhere. That the text, even following the alleged elucidation of the phonetic values of so many symbols, looks like such a mishmash, indicates that nothing has been achieved. Speculations about Mandean from certain contributors look like woo to me, rather than serious linguistics. Best example is Darren Worley's disquisition at http://stephenbax.net/?p=1080. And we have an anonymous Finnish contributor. Why anonymous? It all looks rather like the stuff published from time to time arguing that scratches on American rocks are in fact ogham inscriptions in a weird mixture of languages: this bit looks Irish, that bit looks Basque; here we have Algonquian influence ... and on and on. No coherent meaning is ever elucidated.
So there is some odd material in that website. Now, I even have problems with Bax himself. As I have indicated: in real decipherments, once the stage has been reached where cross referencing produces plausible phonetic values on the scale indicated by Bax, the thing falls into place at once. That happened with Ugaritic and Linear B, and even with hieroglyphs. It's Champollion's triumphant "eureka": Je tiens l'affaire! http://www.bbc.co.uk/history/ancient/egyptians/decipherment_01.shtml
Nothing like that has happened in this case, so I think Bax is not on the right track.
ETA The "eureka" happened with Hittite too. See http://www.sjsu.edu/faculty/watkins/hittite.htm.
Hrozny's "nu ninda en e-iz-za-te-ni wa-a-tar-ma e-ku-ut-te-ni" is not an incomprehensible mishmash of various different language groups, but a clear statement in an Indo European language "Now you will eat bread and drink water". As soon as the correct affinity of the language was postulated the meaning of the text "fairly leaps out" as the article states.
ETA2 And also Palmyrene, the very first ancient language to be deciphered. As soon as it was correctly identified as Semitic, the decipherment was performed "literally overnight". See http://books.google.co.uk/books?id=...rene decipherment literally overnight&f=false.
Last edited:
Delvo
Дэлво Δε&#
Mishmash of multiple languages? How in the world does that have anything to do with anything?!
Not at all. Solving an unknown alphabet and building it up to the point of being able to read & comprehend the language it represents has never worked all-at-once like that, and can't possibly do so. The most comparable case you've brought up, having to be built up from nothing one sound at a time and then still needing some kind of link from the sound system to the rest of the language, was hieroglyphics, and here's the full story on those, of which your link only gives one point near the end:
- In 1802, Silvestre de Sacy & Åkerblad started working out a partial Demotic alphabet using names in common between the Demotic and Greek sections of the Rosetta stone.
- Beginning in 1814, over several years of work during which he was discussing this with Champollion, Young:
- discovered sound values in hieroglyphics for a Greek name
- observed that hieroglyphic phonetic symbols resembled Demotic counterparts
- deduced that they were different ways of writing essentially the same spoken language, and
- deduced that Demotic used some non-phonetic elements.
- In 1822, Champollion compared the Rosetta stone with the Philae obelisk, enabling him to identify another Greek name in hieroglyphics, which enabled him to flesh out most of the rest of the hierglyphic phonetic system using comparisons with Demotic phonetics.
- At this point, the hieroglyphic sound system was almost complete, but there was still no way to attach meanings to any sequence of sounds other than the names that the sounds had come from in the first place. Most hieroglyphic words (the ones that didn't contain the last few unknown phonetic symbols) could be pronounced, but were still gibberish. It would have to wait in that state until the project caught the attention of someone who spoke whatever language it was, or one close enough to it. There's no telling how long that wait could have been.
- Purely by random serendipity, it was almost immediate. An individual who was familiar with Coptic, coincidentally Champollion himself, was already involved, and soon came up with the idea that the sounds represented by ancient Egyptian writings could be related to Coptic sounds for the same meanings, which he used to fill in the last few sound values in the phonetic system and apply Coptic's meanings to ancient Egyptian sounds. The fact that this was the same person who had done stage 3 was a coincidence. Even the involvement of a separate person such as was necessary for stage 4 so soon after stage 3 would have been a coincidence. Stages 1 and 2 had been done by non-Coptophones, and stage 3 very well could have been too (particularly Young was already heading in that direction), since none of that required any knowledge of Coptic; only stage 4 did. If that had been the case, it could very well have been years/decades before someone else came along who happened to have the outside knowledge required for stage 4, but as the dice rolled, it just wasn't.
- Although the first rough version of the hieroglyphic alphabet was finished and its relationship with Coptic was established in 1822, and its use in Egyptian names (not just foreign ones, with Egyptian ones being non-phonetic) was confirmed in 1823, Champollion was still editing & refining his theories about how to read & write it at least as late as 1828 (the date of his last publication on the subject during his life), possibly longer (with two other works being published after 1832, the year of his death, but written I-don't-know-when).
With the Voynich Manusript, you're looking at the counterpart to stages 1 & 2, a process that's inherently limited to getting only as far as about stage 3 (or maybe slightly farther, if the sound system were to be 100% worked out instead of just 95%), and declaring it "nothing" because it's not stage 4. It's not supposed to be stage 4 and nobody involved is claiming it's stage 4. It's just the part that stage 4 depends on getting done first.
Last edited:
I was referring in an exaggerated way to this from your previous post, when you were discussing the underlying language.Mishmash of multiple languages? How in the world does that have anything to do with anything?!
I will respond soon to the other points you make. Thanks for the long and informative post.
@ Delvo Can I ask you to comment further on one of your own observations?
In the light of this pre-existing understanding, Champollion's knowledge of Coptic was believed to qualify him for the attempt at decipherment. See his wiki biography at http://en.m.wikipedia.org/wiki/Jean-François_Champollion
But the circumstance that these inscriptions were being studied by a person with a knowledge of Coptic was in no way a matter of "random serendipity". It was already known, or at least correctly believed, that the ancient Egyptian writing system underlay a language resembling later Coptic.
http://en.m.wikipedia.org/wiki/Decipherment_of_Egyptian_hieroglyphs
In the light of this pre-existing understanding, Champollion's knowledge of Coptic was believed to qualify him for the attempt at decipherment. See his wiki biography at http://en.m.wikipedia.org/wiki/Jean-François_Champollion
How likely is it that this process will yield, not a decoding, but a creation? In other words, are they extracting meaning or injecting it?
Could one algorithmically produce the semblance of a language by an iterative process of "best fit" a la the bible code? How would we know if this were happening and at what step in the process might it become obvious?
Could one algorithmically produce the semblance of a language by an iterative process of "best fit" a la the bible code? How would we know if this were happening and at what step in the process might it become obvious?
Once the correct solution has been found, it is almost always evidently correct, at least in the case of a phonetic script. If the phonetic values have been found, then applying them to a further set of symbols will produce coherent text. It's much the same as recovering an enemy cipher key in war. Once the correct key has been found, the intercepted messages start to make sense. Anyone in possession of the key--NOT solely the original decipherers--can use the key to recover unambiguously the meaning of any further texts, even those not studied in the course of the original decipherment.
This is what I am not sure has happened, or is happening, in the case of the alleged Voynich decipherment.
Last edited:
I'm not impressed by this material from Bax, as anything in the nature of a translation. It likewise illustrates what I mean by "mishmash of languages".
In the passage I cite below we have wooish content expressed in two languages, nay three languages, because "Voynichese" is there, though nothing can be said about it, because it is "ciphered". This isn't much like the decipherment of hieroglyphs.
Given enough imagination I could turn a bunch of squiggles into a recipe for braised nanny goat's liver with coriander if I were allowed to propose that my alleged words were highly abbreviated and culled at will from disparate languages; and if I were in addition permitted, if need be, to state that some of them were in a ciphered unknown tongue. Also the "deciphered" text looks very odd, and a bit crackpotty.
Last edited:
FromBelgiumWithLove
Unregistered
The problem being that there are no further texts to "decipher" (that term presupposes there's some kind of genuine content in there in the first place). There's this one, not very large, book, and nothing else. Which fact, combined with its dating and geographical origin, is exactly what leads many people to think it's probably the work of a single hoaxer/fraudster/crank, pick whatever word you want.
Aepervius
Non credunt, semper verificare
I contest the "not very large" book. It is a 240 pages all fold out counted, and 5 cm thick.
Plenty enough to find coherence. But at the moment *again* as with previous attempt all we have are a few attempt at associating plant phonetic name or sentence part with some language and go no further. The above with plant name is not the first attempt of the same nature.
So color me skeptical, until somebody has consistentely decoded more than a few isolated word, but real paragraphs with meanings.