29-09-2025, 01:20 PM
Thanks for the thoughtful reply, as always interesting!
I totally agree there are many ways that repetition can show up in texts, and surely we could generate even more examples. But as you say, we don’t know the meaning of the Voynich words. I just wanted to lay out some thoughts from the perspective of a non-linguist who is wrestling with transliterations and looking at the text like a data scientist.
Your "I would rather do that, than this thing. This thing that they want..." example is actually quite close to what I was trying to capture in my thread: each Voynich "word" might work like a small grammatical piece that strings together into clauses.
The "called cauliflower, coliflower, calliflower, or kaleflower" case feels less convincing to me here. qokedy / qokeedy / qokeeedy are scattered throughout the manuscript. An author might occasionally list variants, but usually they’d settle on one form (“coliflower”) across the whole text. Or at least is what intuition says to me.
The "sanctus sanctus sanctissimus..." type of repetition makes sense, and maybe the e / ee / eee alternation could indicate superlatives or inflections. My hesitation is that here a single character ("e") would have to cover multiple different grammatical roles, which feels a bit messy or hard to justify from a usability standpoint.
Numerals — yes, they could definitely be part of it, or formulas involving numbers. But then I’d expect the illustrations to play a stronger role in signaling that, and they don’t always match up (I know, that’s subjective).
And about "many similar words": in our case it’s really just a single character repeated 0, 1, 2 or 3 times. That makes them very similar words indeed.
I suspect these kind of repetitions have not been found as they would be a leading clue to solve the mistery.
Anyway, my post was mainly meant as a reflection on one of the toughest passages, where most (if not all) translation attempts break down: those three lines packed with qokedy, qokeedy, etc.
I thought it could be interesting to try a bit of "reverse engineering" to throw out a hypothesis starting from these lines themselves, rather than ending up stuck with them at the end of a possible solution and having no way to explain them.
As you said in a recent post, the point is to make hypotheses and then see whether they hold up or not. My intention was simply to suggest that maybe the biggest stumbling block is a good place to start pulling the thread.
I totally agree there are many ways that repetition can show up in texts, and surely we could generate even more examples. But as you say, we don’t know the meaning of the Voynich words. I just wanted to lay out some thoughts from the perspective of a non-linguist who is wrestling with transliterations and looking at the text like a data scientist.
Your "I would rather do that, than this thing. This thing that they want..." example is actually quite close to what I was trying to capture in my thread: each Voynich "word" might work like a small grammatical piece that strings together into clauses.
The "called cauliflower, coliflower, calliflower, or kaleflower" case feels less convincing to me here. qokedy / qokeedy / qokeeedy are scattered throughout the manuscript. An author might occasionally list variants, but usually they’d settle on one form (“coliflower”) across the whole text. Or at least is what intuition says to me.
The "sanctus sanctus sanctissimus..." type of repetition makes sense, and maybe the e / ee / eee alternation could indicate superlatives or inflections. My hesitation is that here a single character ("e") would have to cover multiple different grammatical roles, which feels a bit messy or hard to justify from a usability standpoint.
Numerals — yes, they could definitely be part of it, or formulas involving numbers. But then I’d expect the illustrations to play a stronger role in signaling that, and they don’t always match up (I know, that’s subjective).
And about "many similar words": in our case it’s really just a single character repeated 0, 1, 2 or 3 times. That makes them very similar words indeed.
(29-09-2025, 11:44 AM)Jorge_Stolfi wrote: You are not allowed to view links. Register or Login to view.So it would be prudent to first define mathematically what would constitute a "repetitious phrase", and then test whether such a pattern also occurs in other comparable texts, such as "alchemical herbals" and medical treatises of the early 1400s.
I suspect these kind of repetitions have not been found as they would be a leading clue to solve the mistery.
Anyway, my post was mainly meant as a reflection on one of the toughest passages, where most (if not all) translation attempts break down: those three lines packed with qokedy, qokeedy, etc.
I thought it could be interesting to try a bit of "reverse engineering" to throw out a hypothesis starting from these lines themselves, rather than ending up stuck with them at the end of a possible solution and having no way to explain them.
As you said in a recent post, the point is to make hypotheses and then see whether they hold up or not. My intention was simply to suggest that maybe the biggest stumbling block is a good place to start pulling the thread.