Just noticed that the image link isn’t working in the old thread. New link: Dropbox - File Deleted
Here we go again. (As seen here, but I thought it would get more traction in this thread.)
Their test decipherment of the first line doesn’t inspire confidence: “She made recommendations to the priest, man of the house and me and people.” 
Ah…so human fantasy interpretation is a major factor?
Yeah, I saw that a couple of days ago. My reaction was the same as yours.
I suppose if I wanted to get my name in the news, all I have to do is claim to have cracked the Voynich manuscript. Either that or I’ve invented a new energy technology that will save the world. Or I have a cancer cure. Or some new astronomical phenomenon is evidence of aliens.
Yoynich manuscript reveals the fate of Amelia Earhart!
Who is more to blame, all the Bible-Code-type cranks, or the media H. L. Mencken was complaining about 100 years ago? Same as it ever was.
But…but…but!:
What an extraordinary breakthrough!:smack:
I’ll bet it also includes the words for “plant,” “water,” “earth,” and “woman.”
Yeah, that part reminded me of those scenes in movies and TV (I’m looking at you, Wargames!) where a computer is breaking a long password one digit at a time. As if there was a way to know the correct digit for each position independently. (So, say, a 10-digit code would go from needing up to 10,000,000,000 tries to needing up to 100 tries.)
Another view on the latest AI decoding attempt. Verdict: failure.
“AI didn’t decode the cryptic Voynich manuscript — it just added to the mystery”
A Nov 2016 New Yorker article summarizes the basics of the story. I see that Voynich bought the manuscript in 1912: a 17th century letter was attached which claimed that the book had been “sold to Holy Roman Emperor Rudolf II at a reported price of 600 ducats and that it was believed to be a work by Roger Bacon.”
A Venetian ducat contained 3.545 grams of gold; 600 ducats worth of gold would be worth about $90,000 today. So there was incentive methinks to create a fake. Assuming the letter was accurate. Which it may not be. Because I understand the Roger Bacon authorship has been debunked.
Puzzle: there are all manner of medieval fakes. But this is the only example I know of a fake book in an imaginary language. And in terms of letter frequency, it’s apparently a plausible fake.
Greg Kondrak has been a friend and colleague of mine for almost 20 years, and I can assure you that he’s no crank, nor is he much of a publicity seeker. He’s just a hard-working scientist who enjoys testing out state-of-the-art techniques from artificial intelligence on famously unsolved linguistics problems. This sort of thing isn’t unusual in my discipline. Most approaches run into dead ends, but there are occasional breakthroughs. A notable recent example from our community is Kevin Knight’s successful decipherment of the Copiale cipher, which used techniques inspired by statistical machine translation.
I was at the talk (at last year’s annual conference of the Association for Computational Linguistics) where Greg first presented his work on the Voynich decipherment. It was a typical staid scientific presentation, though that didn’t stop people like me from sitting on the edges of their seats in anticipation of a big revelation. (I was secretly hoping for one like Andrew Wiles made when, at the conclusion of an unassuming three-day lecture, he matter-of-factly observed that he had solved Fermat’s last theorem.) Alas, it was not to be—Greg is not known for being sensationalistic, and was quite upfront that he had not deciphered the text. Using the hedging language beloved of all working scientists, he claimed only that he had found good evidence (but not hard proof) that the manuscript is written in Hebrew, or a closely related language, with the vowels removed and the consonants rearranged within words. In the public discussion after the talk, I asked him why anyone would mangle a text in this particular manner, and IIRC he was able to point to other texts that have been anagrammed in this way. So his hypothesis is at least plausible.
I’m pretty sure that there were journalists in attendance at the conference, since I got advance requests from one to be interviewed about my own work. It could be that’s how word of his study got to the media. But regardless how the press got involved, I think you can blame any sensationalization of the work on them rather than on Greg and his coauthors. Their job is to sell newspapers (or eyeballs to advertising-supported websites), and you must admit that grandiose headlines about some obscure professor cracking a 500-year-old mystery will generate a heck of a lot more subscriptions and clickthroughs than repeating only the vague and tentative findings in Greg’s original paper.
As someone who is active on the Unresolved Mysteries Reddit and thus has seen it come up many times: I think the V.M. is bullshit.
The “writing” looks like “kinda-sorta-Latin-Alphabet-but-not-really.” I.e., made to look just similar enough to real characters (that would be recognizable to Europeans) as to be believable, but also different enough that they’d look “exotic”, without actually meaning anything. Also, the “character” that keeps appearing, consisting of two vertical parallel lines joined by a loop at the top, is something that I used to randomly doodle on stuff when I was in elementary school, basically because it’s a simple yet satisfying motion to make with a pen and it looks cool. I suspect the author of this “manuscript” did so for the same reason.
As an aside, I can’t be the only one who sees the word “golf” all over the pages. LOL, this thing is such a fucking hoax, such a troll job. We’re continuing to feed the troll by talking about it!
Pretty sure that troll is long dead.
Thank you for your comments and personal perspective on this! To be sure, I had not in mind to cast any aspersions upon Kondrak. Advanced statistical and AI techniques are sine quibus non underlying any serious analysis of such cipher manuscripts, though not always the end-all and be-all; I have said as much myself.
Your story proves the facts in this case are worse than I assumed. It’s one thing if I leak to various journalists and confidently blag in subsequent interviews my remarkable translation of the Voynich MS from Venusian, relying on the mediator’s lack of basic fact-checking. It is quite worse when reporters hang about large conferences trying to create sensational news headlines when there are none.
My first reaction to your post is, WTF?
My second reaction is, is English your first language, and if so, WTF?
More succinctly, then:
-
I never meant to slander Greg Kondrak and apologize if my comment was interpreted that way.
-
Reporters who are quick to publish sensational stories without due diligence are ultimately doing themselves and the profession a disservice.
I believe I am able to find the words “warning” and 'banned" in the text. Have no idea what it means, however.
I’ve always suspected that the manuscript starts out, “If you can read this, you can get a good job and make lots of money!”
I’m not a VM scholar, but I would never misunderestimate the ability of people in any era to allow themselves to be fooled, nor the ability of fairly clever people to provide just the materials that will sell for the most money. A (forged) manuscript like this would have been similar to a holy relic, possibly occupying a place of great prestige in somebody’s private library. As for the complexity of it and the time it would take to create it…keep in mind that the forger might have anticipated selling it for the equivalent of many years of an average person’s wages. Vellum? Takes money to make money. Even then, investors in phony schemes were not unknown.
IMHO, this continues to be the simplest solution. The people of that time were ignorant of advanced technology, not stupid.