Cvetka's Slovenian Theory Thread

[tavie]

1. Translation of the VM should be the last step, not the first step.  Many of those who boasted about the translation of a page or even a sentence, have been discredited right away. The Stična Codex, written about the same time as the VM (between 1428 and 1440) in Slovenian language in Latin letters, had been studied by many linguists before the translation was accepted. Hopefully I will acquire a group of Slovenian linguists who will tackle this. For now, I am very grateful to Dr. Alenka Čuš, for her advice on Slovenian language, as well as to Dr. Duša Krnel Umek for her expert advice on Slovenian ethnology.

No one is asking for a complete translation, or even a polished partial one.  If it is ever solved, no doubt the solution will contain mistakes that are eventually ironed out by experts in the language. This is what happened with decipherments in the past.  But no Voynich theory will ever be viewed as credible based on only identifications of individual words. 

In particular, you mention grammar on your website and here.  But how can the grammar of your solution be tested without having sentences to analyse to see if the grammar corresponds to medieval Slovenian?

Many of those who boasted about the translation of a page or even a sentence, have been discredited right away. 

That's exactly my point.  These theories were discredited precisely because they were put to the test and they failed the test.   Working through sentences is how to test a theory.  What you have said here makes it sound as if you want your theory to avoid being tested.  Fair enough if it's because you want to wait until the theory is more developed and robust...but then why promote it and say you are certain about your identifications if it's not ready for testing?

I'm repeating myself from earlier but:  it is incredibly easy to find meanings of Voynich words in isolation without context.  That's why the first hurdle for a theory is to provide the words in context, aka the full sentences.  The aim is to show that the proposed identification is consistent with the target language's grammar and most importantly...that it makes sense.  For the few theorists who have tried to tackle this hurdle, the result is always the same:  a word salad that makes no sense in English, and likely neither in the target language.  And that's why the second hurdle - translating the full context of those sentences via their paragraph/page - is seldom attempted.

But many don't even tackle the first hurdle of producing even a sentence.  The reason is because they find their individual word identifications don't make sense when put together in a sentence.  And then they rationalize this away by telling themselves "of course my discoveries are still correct; it's just that experts in the language are needed to put them all together".

I might be unfairly mischaracterizing your approach, but it sounds awfully reminiscent of the above.  It sounds like you are saying here and on your site that while you have sufficient expertise for your identification of individual words to be certain, experts in medieval Slovenian are needed to read full sentences.  If you are finding it hard to apply the system to full sentences, you should consider that it is not your knowledge of medieval Slovenian that it is at fault:  it is the system. 

Again, the crux of the issue is:  full sentences and paragraphs are needed in order for a theory to gain credibility.  This is the only way to demonstrate the system produces meaningful text consistent with both the target language's patterns and Voynichese's quirks.  Any identifications of individual words, no matter how many, could always be coincidence.  And not recognizing this sets you up for confirmation bias.  You say theories have been "discredited" by early translation, but nothing discredits a theory more than confirmation bias. 

I'll go through the first main findings in your article to illustrate this and the point I was making with #3 about being certain far too soon.