• Options
• Login
• Desktop Style

Polyglot

• Index

  • The Voynich Ninja
  • Voynich Research
  • Analysis of the text
  • Polyglot
• Polyglot

• Polyglot

  -JKP- > 02-04-2019, 08:06 AM

  It's wise to be careful about VMS "translations" that rely on an assumption of "polyglot" in order to be meaningful. The reason is simple... with enough degrees of freedom even nonsense text can be made readable (or almost readable).
  
  But... I thought it might be worth discussing this subject because some languages are inherently polyglot.
  
  
  For example, I was just reading a 14th-century letter in Frisian, and it is such a weird experience because it's English, Norse, Norse, English, Norse, German, English, Norse, Norse, Norse, English, English, German... It's a blend of these languages and if you know all three, you can pretty much read it.
  
  I did notice that the grammar leaned more toward English than Norse (which in itself is interesting because Old English and Middle English sometimes have Norse grammar mixed in, but only bits of it, here or there, not the whole thing).
  
  
  So, if Frisian were enciphered and someone were looking for word or letter-frequency patterns, it might still be possible to decipher a relatively simple substitution code, but... it would be a challenge, because it mixes three languages with different spelling structures.
  
  
  In some ways, English is polyglot as well. A huge proportion is French, words like chauffeur, entrepreneur, bureau, critique, beau, recipe, bourgeois, valet, hors d'oeuvre, depot, chamois, ballet, armoire, cafe, a la mode, cliche, decor, savant, fiance, etc.
  
  The "eau" combination is common in French but not otherwise characteristic of English words. Patterns like this could throw off computational attacks because they would not be consistent or in the same frequency as in the parent language.

• RE: Polyglot

  Koen G > 02-04-2019, 08:27 AM

  I wouldn't call Middle Frysian inherently polyglot. It's a language in its own right, but because of where it sits in the family tree it shares properties with different languages.
  
  Now English, that's a different story because it integrates a huge number of French roots into a Germanic system. Take your sentence, where I'll put non-Germanic words in bold:
  
  

  Quote:I thought it might be worth discussing this subject because some languages are inherently polyglot.

  
  Now if the language underlying the VM is had something similar happen to it as English after the year 1066, then it might be inherently polyglot, and it would not be strange to mix roots from two or three language families.
  
  Even in such a case, I would expect some structure in the polyglot mess. For example, English retained most of its basic, everyday words, grammatical words, pronouns... From your sentence: I, it, be, this, some, are... Such basics are entirely missing from most (if not all) VM solutions.

• RE: Polyglot

  davidjackson > 02-04-2019, 08:42 AM

  Any polyglot solution has to have a grounding in historical reality. The languages have to be contemporaneous and ideally geographically linked, as in your Old English / Norse example. It has to have been possible for someone to naturally mix the languages in their day to day life. These sort of polyglot pidgins are perfectly naturaly, especially in border zones or recently conquered areas.
  Trying to mix, say, Arabic and medieval French (editor's note: what about Jerusulem during the Christian crusaders occupation? They probably spoke it then)... OK, Occitane and Norse, is a bit of a no-no for the simple reason that nobody in history ever spoke like that daily in real life, even if they knew the two languages. Whereas it is perfectly reasonable to assume that a pidgin of Norse / English / German could have been spoken (although taking it as far as a creole is probably is step too far!).
  Mind you, I dread to think what sort of hoity-toity English would have enough French loanwords in it to throw off the computational attack... my fiancée is most desolée today because the chauffeur went to a critiquer the ballet, and took the valet with him because, of course, they are so savant about these sorts of things, although I suspect they may both be in the armoire if you understand what I mean by the cliche; but it meant the entrepreneur who was bringing the recipe for the hors d'oeuvre, so á la mode you know if a little bourgeois, had to take the bus and got lost in the depot whilst buying a café because they were admiring the decor...

• RE: Polyglot

  -JKP- > 02-04-2019, 08:56 AM

  David, I laughed so hard, I had to put my cup down so I wouldn't spill.

• RE: Polyglot

  Koen G > 02-04-2019, 09:27 AM

  We have the luxury that English and its immediate parent languages are very well attested. But what of similar situations where there is very little written record? 
  
  Imagine for a moment that we knew only the English from before the Norman conquest, but the VM turned out to be post-Norman English. Depending on the type of document, around 50% of English vocabulary is of Latin/French origin. We'd be confused by the way Latin and Germanic roots are mixed.
  
  What David says about historical background is of course crucial. But it may be difficult to eliminate possibilities. Latin, Greek and Arabic are all tied to sciences and had international roles. So if someone said that the VM mixes Latin, Greek, Arabic and a fourth language of choice, you'd have to accept the possibility.
  
  The problem is not the mixing in itself but rather the way it's mixed. Specialized vocabulary is fine, but there must be some structure. In a regular sentence you can't just allow yourself to pick at random. Even in a language like English, that has been severely exposed to Latin influences, the core vocabulary remains almost entirely Germanic!

• RE: Polyglot

  davidjackson > 02-04-2019, 09:51 AM

  As I said elsewhere recently: Grammar, grammar, grammar !
• Next Oldest Next Newest