Re: Mystery Phoneme

>In contrast to a normal /t/ which is made with the tip of the tonge above
>and behind the upper teeth, the mystery sound is produced by bracing the
>tip of the tongue behind the lower teeth and raising the middle of the
>tongue to the same point that the tip would touch in a normal /t/ (or maybe
>slightly further back). The resulting sound is softer than /t/, and sounds
>slightly different in spoken speech. It is still a voiceles stop,
>articulated in roughly the same place as /t/, but with a different part of
>the tongue. Is there such a thing, or is there any reason to believe that a
>human natlang might have this sound as distinct from /t/ (or is it even a
>case of ANADEW?). Has anyone else used it in a conlang?
This is how I pronounce my alveolars when I speak Danish, but I can't
certify that it is how Danes are pronouncing their alveolars.  I learned by
experience that I could best mimic the sound of spoken Danish if I put the
tip of my tongue down behind the bottom teeth and used the tongue blade as
the articulator.

The two sounds are not contrastive in Danish.  (I'm not certain that Danish
even has the more normal type of alveolar.)  Is there enough contrast
between the two sounds for them to be used as separate phonemes?  Based on
my experience with Danish, I will venture a guess that there are things
that you can do to the pronounciation of the new sound that you've
discovered that would make it a candidate for a phoneme to contrast with [t].

>Lacking better terminology, I'm calling it an "antiretroflex t" (retroflexs
>curl the tongue upward, in this sound, the tongue curls downward). Come to
>think of it, you could also have a whole series of "antiretroflexes",
>including an "antiretroflex" d and n.
I would call it "laminal" rather than "antiretroflex."  It is laminal as
opposed to apical, because the tongue body, rather than the tip, is what
contacts the alveolar ridge.  And yes, you can make other alveolars in this
fashion.  I swear this has gotta be what Danes do to most of their /d/'s,
and it sounds great if you turn it into an approximate in a postvocalic
environment, which is what I think is happening in Danish.  It's just the
weirdest sound, and I love it.  It sounds an awful lot like [D], but it's
different enough that you can tell that it definitely isn't [D].  Get a
Dane (if you are ever so lucky) to say "marmalade" (jam) for you, and
listen to the intervocalic /d/.  (And while you're at it, get him/her to
say "Turkiet" (Turkey), and hear what happens to the final /t/.)

Isidora