Having browsed Stephen Bax' paper "A proposed partial decoding of the Voynich script" (available here), as a scientist in the natural sciences (physics, mathematics), I find his proposed decoding of the Voynich Manuscript plausible. However, I have no means of judging whether the evidence is strong enough. To me, it seems both plausible and too tenuous at the same time.
So, what is the linguistics community's view on Bax' theory?
I remember finding quite a few WTF moments in his paper, but it was a year ago, so to refresh my memory, I opened the paper again, and skimmed through. For example, on page 24, he compares word "coriander" in different languages and tries to compare the phonemes, and, among other things, compares ending -ον in κόριον to -n- in English "coriander", which are not related at all, as -ον in κόριον is simply a nominative/accusative singular ending which is not manifested anywhere else (dative κορίῷ) -- this is something pseudoscientific folks love to do, compare random morphemes to random morphemes. Not to say 1) he literally leaves it as KORIJADANA as if not knowing how Linear Greek works (and even reconstructs a vowel from -A- in the syllable -DAN-, which is actually just a quirk of the Linear Greek alphabet which was syllabic) 2) κορίαδνα is plural, as mostly plural was used for coriander in Ancient Greek, and for some reason, it's compared to singular κόριον (which kind of makes sense, otherwise there would be no -ν- in κόρια for his comparisons). He also says that 'coriander' in the mentioned languages derives from Greek κόριον, which is nonsense, as κόριον itself is a shortened version of κορίαννον, and most of the mentioned European words actually borrow the word from Latin coriandrum or Greek κορίανδρον (another variant). His table also compares Modern English coriander (with lost endings) to ancient tongues directly.
Such strange things here and there suggest me the author is not totally sure what he's doing. On almost every page, there are clueless comparisons of random things to random things with argumentation typical of amateurish linguists the internet is full of. For example, on page 44, he compares Chiron to Turkic kara to Russian чёрный (he found all those examples on Omniglot!). How cool is that? The typical (amateurish and unscientific) argumentation of his is "let's find similarly looking words via Omniglot/Google in different random languages; random alternations like L/R or K/KH/CH don't matter, as for vowel inconsistences, we can explain it via abjad. When a pattern is found, for example K/KH/CH+vowel+R/L (this template works for roots: khal, kir, chur, etc.) while comparing Sanskrit to modern Russian to Sumerian, we can suggest it's a word in the language of the Voynich manuscript as well. We have a vast material of whole 10 words decoded this way."
