Dangeur! Diphthongues inattendues!

In Brittany last week, home of NECESSARY BABIES (of which we had some once), and tourist bumph that recommended a visit to St Flavour's (St Saveur), and somewhere where we could see an extraordinary triomphe l'oeil painting (a bit like The Fighting Temeraire, I suppose  –  but it was raining so we didn't find out) I spotted what I assumed was a typo at the foot of a carte de vins. (I still think it was; I'm just covering myself with that assumed):

Attention: l'abus de l'alcool 

est dangeureux pour la santé.... etc

I thought little of it at the time, dismissing it as just another semi-literate typo; or, conceivably – if improbably – there was a thitherto unknown pair of words: dangéreux and dangeureux (the second having some specialist application – perhaps in official pronouncements). More likely, though, as MrsK said at the time, it was just the sort of spelling mistake that people make – even foreigners. (She also questioned my spelling – but I am a forgiving man.)

The following evening, though, I saw it on another carte de vins – reinforcing, to my chagrin, the official pronouncement hypothesis. On further reflection I found two other possible explanations, both with an interesting linguistic foundation.

1. Both cartes took their spelling unquestioningly from an official source (that just happened to have a typo in it). This would be reminiscent of the way students of philology can trace the provenance of a manuscript through their accumulation of  traits in successive generations. I mentioned one such manuscript-based linguistic happenstance in an earlier post about bald owls BATS (not exactly the same, but similarly depending on a chance mistake, concretized in subsequent usage).
2. A really interesting possibility, based on the phonology of Breton. A glance at  a map of the region will show several examples of place-names (the last refuge  of etymological nuggets) such as Saint Domineuc –
   
   
   
   – with the digraph eu where one would expect an i. Perhaps one could extend this to apply to vowels such as é (also a front vowel – one produced towards the front of the mouth [as opposed to back vowels such as a, o, or u]). Is Breton phonology characterized by a tendency to substitute  eu for a front vowel?*
   
   In that case, the dangeureux typo would be more likely to occur in Brittany than elsewhere in France – a possibility that is intriguing (if unlikely to succumb to further research, given the state of the hedge [which is in urgent need of a haircut]) 

b
Update: 2016.07.11.22:45 – Added clarificatory parenthesis in red.

PS And a trilingual crossword clue (a  new invention – the clue is followed by its character count and indications of three languages, the first two representing languages used in the clue, and the last indicating the language used in the answer)

Expression of gratitude interrupted just before the end by article reversing, for example, Ash. (8,  Fr – D – Fr)

Update: 2017.01.06.11:15 –  Added PPS

PPS: Answer: MERCREDI (Came to me in a boulangerie [with an horaire {mot juste?} showing opening times]).

Update: 2018.03.31.12:35 –  Added PPPS

PPPS: Added explanatory parenthesis in blue to PPS. Without it, the detail was pointless – just an old man's memory (which it was, of course; but my point was that seeing the word was the stimulus for the cruciverborhoea [and don‘t bother looking that up 😐] )

Update: 2018.10.12.16:15 –  Added footnote.

* This lumping together of e and i for special treatment is not random. As an example , consider what happens tn to a c in Spanish or Italian when it precedes an e or an i. The sound becomes /𝜃/ or /s/ in the former case. and /ʧ/ in the latter. (This does not involve the modification of the vowel sound itself – my argument is just about anything [any linguistic change, that is] that happens only to e and i and not to other vowels.)