Re: [embonpoint] was once a completely positive term in France

 Ar an t-ochtú lá is fiche de mí Deireadh Fómhair, scríobh Peter Moylan:

 > [...] When it's someone speaking Irish, an extra factor comes in: my
 > vocabulary is so limited, and my command of Irish spelling so poor, that I'm
 > struggling to understand anything at all. Under those conditions, I can fail
 > to distinguish two words even though their pronunciation is different.

OK, so no deep-rooted lack of perception, “just” a deficit in practice.

 > There's also the fact that recognising an accent does not imply being
 > able to analyse the features of the words being spoken. I used to live
 > in Melbourne, at a time when it had many recent immigrants, and when I
 > was in a crowd -- on a railway station, for example -- it amused me to
 > guess which languages people were speaking. I think those guesses would
 > have been very accurate. These were languages that I didn't speak or
 > understand, but I could pick them because different languages have
 > different rhythms and dominant sounds, and one can respond to that
 > without knowing what any of the words mean. A lot of what registers is
 > subconscious.

Yeah, I get you, but I do think this can be leveraged to pick up on phonemic
distinctions when learning another language.

 > Here's another example. I once got lost in central Paris at midnight, so
 > I stopped a passer-by and asked for directions. He told me where to go,
 > I thanked him, and we went in our different directions. It wasn't until
 > I had walked a whole block more that it suddenly hit me that that man
 > had been speaking French with an Australian accent. The recognition was
 > in my head, but it hadn't come to the surface. And he, presumably,
 > hadn't noticed that I was an English speaker.

Clearly the French of both of you was good enough for the task to hand, no bad
thing.

--
‘As I sat looking up at the Guinness ad, I could never figure out /
How your man stayed up on the surfboard after fourteen pints of stout’
(C. Moore)