Voynich Reconsidered

[kckluge]

[quote='Anton' pid='53906' dateline='1677068304']
[quote]
What I am suggesting is that hypothesizing something like the text being "a forgotten language [in] an unknown script designed specifically for ethnic group having no their script of their own" is all to often used as an excuse to ignore, dismiss, or otherwise fail/refuse to engage with the actual, concrete statistical properties of the text. It isn't.
[/quote]

Well, I would say that in the first place such hypothesizing goes ahead of even considering the statistical properties of the text. We have seen many such examples.
[/quote]

We are in vigorous agreement on that point -- statistical properties are necessarily statistics computed over some specific considered body or bodies of text. Suggestions of additional bodies/types of texts to examine should be welcomed.

[quote='Anton' pid='53906' dateline='1677068304']
[quote]
E.g. late prof. Bax (who was not some enthusiast amateur but a prominent linguist and whose 2014 article first attracted my attention to the VMS) did not seem to have considered statistical properties when he issued the said paper - and he was much criticized for that by the VMS-research veterans. I am far from linguistics and do not know whether statistics are or are not the second nature of a modern linguist. I think there is computational linguistics which revolves around statistics, but probably there are other flavours of linguistics which are statistics-unaware.
[/quote]

Be that as it may, investing some reasonable level of due diligence in a literature search and engaging with key relevant prior work is an expected part of doing and publishing research. The goal here should be for everyone to broaden their sandboxes, not bury their heads in the one they're accustomed to.

In hindsight, I probably didn't make the point(s) I was trying to make as clearly as I would have liked at the start, and I appreciate the interaction because that helped me recognize that and try to hone what I'm trying to say. To recapitulate/summarize in bullet form:

• People coming up with unusual/unconventional forms of natural language text to consider (or, for that matter, alternatives to the "text" being text at all) is a good thing
  
• Given that linguists have specialized knowledge of languages, their helping come up with such unusual/unconventional possibilities is a very good thing
  
• Given that linguists also have knowledge of/access to specialized analytic methods & tools from their field, their applying those to the text of the mss. is also a very good thing
  
• Having said that, however, the text of the mss. is still what it is...
  • Cipher types aren't *ignoring* the possibility that the mss. text is in a natural language rather than a cipher, they are *explicitly rejecting* it (with a narrow set of known exceptions including abjads & devowelled text)
    
  • They are doing so for perfectly sound evidence-based reasons
    
  • This is a consequence of the equivalence between an unenciphered text in some language in an alphabetic script and the same text/script having been enciphered with a Caesar Cipher with shift = 0 -- all the arguments against the text being a (single glyph for single letter) substitution cipher apply
    
  • Those reasons don't go away simply because someone is approaching the text from a different disciplinary framework; they have to be engaged with
    
  • For that matter, it's entirely possible to find linguists who are of the same opinion (whether for similar or different reasons)
    
• With regard to possibilities that fall outside the body of texts used in making those arguments, as always the burden of proof is on the person making the positive claim
  • Anyone disputing the rejection of the "natural language text" hypothesis needs to produce a corpus whose characteristics are consistent with those of the mss. text
    
  • Making vague handwavy arguments about creoles or phonetic spellings or whatever as if that's a "get out of jail free" card doesn't cut it
    
  • That's not me being a jerk, that's how science works
    
  • (Nor for that matter is that me being a naive falsificationist; I always liked Allen Newell's view that hypotheses are like graduate students: you don't abandon them at the first signs of trouble, you try to find ways to fix them)
    

Karl