There are reasons to be very wary of trying to prove connections linguistically. The
Popol Vuh (and any other colonial sources in Maya) are very tricky because of the complexity of the language particularly the glottal stops, because the documents often do not record these. The same problem (to a lesser degree) occurs in Nahuatl sources that don't distinguish between short and long vowels, which change the meaning of words. Much of the sheer inanity of Clyde Winter's claims of relationships between Mande and Maya arises from his simple-minded linguistic "comparisons" between the languages. It had never penetrated his understanding that a glottal stop is a consonant. (google this dispute sometime
). Here is a quote from Christenson's book.
Allen J. Christenson. 2003.
Popol Vuh The Sacred Book of the Maya Winchester, U.K.: O Books
p. 53 "The
Popol Vuh manuscript however does not consistently use the Parra glottalized letters for the palatal and uvular consonants. In most cases it either ignores glottalization altogether, representing them with the letter c, or uses the letter q to represent glottalized forms. As a result, the single letter ‘c” may have four equally plausible readings.
For example, the word
cac might be read as
kak (their gourd);
kaq (red);
kaq (their peccary);
kak’ (their turkey);
qak (our gourd);
qak’ (our turkey);
qaq (our peccary);
kaq’ (their tongues);
qaq’ (our tongues);
k’ak’ (new);
k’aq (to throw);
q’ak (flea); or
q’aq’ (fire). Four other combinations are possible which have no known meaning in modern usage, but which might have existed in the archaic language of the sixteenth century."
You can see that any translation involves an enormous amount of translator's discretion and guesswork, and in just in this example trying to draw genetic comparisons between "cac" and a word "cac" in any other language supposedly meaning, "gourd" for example, would be on very shaky grounds without other supporting evidence.
Bernard