Re: YAEPT: track

Quoting Benct Philip Jonsson <bpjonsson@...>:

> I once read a book aimed at enabling students to
> make *very* narrow transcriptions of English speech.
> It claimed that the actual phonetic realization of
> '/tSr/' where it occurs is [t`s`] -- not unreasonably
> if the realization of /r/ is retroflex, which it
> apparently is in most accents in both the US and
> the UK.
>
> FWIW [t`], [t`s`] and [tr\`] are all allophones
> of a single phoneme (with an aspirated and sometimes
> a voiced counterpart) in Lhasa Tibetan, all descended
> from *tr if Tibetan orthography is something to go by,
> so this is probably a very natural assimilatory tendency.
I somewhere read that Chilean Spanish has [t`] for initial /tr/. Can anyone
confirm/deny?

> Swedish like English has very various realizations of
> /r/ in different lects.  I don't know if those who
> have [z`] for /r/ have [d`z`] and [t`s`] for /dr/
> and /tr/, but it wouldn't surprise me.
I've got a [z`] allophone for /r/, but it doesn't show up in this particular
position - the clusters are [tr`_0] and [dr`], the flaps/taps turning to trills
in careful speech.

However, I've seen [tz`] in transcriptions of Mälaren Valley regiolect - I
strongly suspect this is sloppy transcription for [t`s`] (I'm tempted to write
[tz`_0], because my [z`] does, for whatever reason, not have exactly the same
point of articulation as [s`] (for /rs/), but I shouldn't assume this is
necessarily more than an idiosyncrasy of mine).

> Interestingly there is a tendency among young,
> mostly female, Swedish speakers to insert a
> [@]-ish sound in Cr  and Cv combinations,
> so that you get [t_h@r\e:] for _tre_ '3' and
> [t_h@v\o:] for _två_ '2'.  It surely is most
> noticeable in those two words, but probably only
> because they are often said in isolation --
> ("How many potatoes do you want?").  Maybe
> Swedish won't have any Cr and Cv combinations
> a hundred years from now!
Daniel Andreasson wrote some posts about this phenomenon years ago. However, he
transcribed the sound as [e], eg. [t_he'r\e:].

                                          Andreas

Replies