Re: [phoNet] Assimilation of English consonants

--- In phoNet@egroups.com, "Piotr Gasiorowski" <gpiotr@i...> wrote:

>
> ----- Original Message -----
> From: Juozas Rimas
> To: phoNet@egroups.com
> Sent: Sunday, April 16, 2000 10:31 AM
> Subject: [phoNet] Assimilation of English consonants
>
> Thereis no inconsistency here, just different preferences.

Different languages permit different kinds of assimilation. English
has no (or very little) voicing assimilation, which is obligatory in
Lithuanian, Russian or Polish (I have ryba 'fish' with [b], but rybka
'little fish' with [p], as well as devoicing before a pause). This is
why speakers of these language have to deliberately practise the
pronunciation of English words like anecdote /kd/, phases like
backbone /k#b/ "minimal pairs" like dog : dock, etc., until they
"unlearn" their native habits while speaking English.

>

A rare exception to the lack of voicing assimilation is width
pronounced as if it were witth.

>
> On the other hand, English consonants (especially the alveolar

ones, articulated with the tip of the tongue) easily assimilate to
the
PLACE of articulation of the following consonant. The pronunciation
of
bad boy as "bab boy" or good girl as "goog girl" can frequently be
heard in casual speech. This assimilation is facilitated by the fact
that English stops are normally unreleased (there is no "puff" of
breath) before another stop, so the acoustic contrast between them is
weakened in this position. A speaker of British English will
typically
pronounce the /n/ in one thirty as a dental nasal, assimilating it to
the following dental fricative [ Î¸], but in one reason the same
/n/
will be postalveolar (with the tip of the tongue retracted or curled
back), assimilating to the following /r/, which is pronounced as a
postalveolar liquid.

>

I also assimilate the nasal before a dental, but not before r.

>
> The palatalisation of alveolar stops and fricatives before "y"

/j/, "sh" /  Ê'£/, "ch"/  Ê§/, "j" /  Ê¤/ (palatal or
palato-alveolar
consonants) is one of the most widespread kinds of assimilation in
English; it affects /s, z, t, d/, which change into /  Ê'£,   Ê'²,
  Ê§,   Ê¤/
respectively:

>
> this year > "thish year" [ Ã°ÉªÊ'£'j ÉªÉ'¹] or

[ Ã°Éª' Ê'£ÉªÉ'¹]

> horse-shoe > "horsh-shoe" or even "horshoe"
> would you like > ['w  Ê'ªÊ¤É'¹'la  Éªk]
> nice journey > ['na  ÉªÊ'£'  Ê¤É'¼n  Éª]
> is George > [  ÉªÊ'²'  Ê¤É'´:  Ê¤]
>

Interesting. I can't recall ever hearing s or z assimilate before tS
or dZ, except possibly late at night in a bar or at a party. Is this
a
British thing? (((apologies, by the way, for the non-IPA: my browser
[Netscape 3.x for Windows 3.3] refuses to recognize Unicode and I
haven't decoded all the character sequences yet; but I may have a new
computer soon which can run newer sofware)))

>
> without your help > [w ÉªÃ°'a Ê'ªÊ§É'´:'help] or

[w ÉªÃ°'a Ê'ªÊ§É'¹'help]
I've heard this occasionally. In my own speech, the t turns into a
"weak glottal stop", so there's no alveolar to be palatalized.

>
> Sometimes American and British English have different preferences

with regard to palatal assimilation, e.g. Parisian usually has [zj]
in
RP (the standard British accent) but [ Ê'²] in General American.

>
> In its early history Lithuanian also had palatal assimilations

affecting combinations of consonants with /j/; these palatalisations
have become "fossilised" as morphological alternations though they
are
no longer active phonetic processes:

>
> pavyd Ä'·ti 'to envy' : pavyd Å¾iu (1 sg.), with d Å¾

[ Ê¤] <
Proto-Baltic dj

>
> Piotr
>
>
>

----------------------------------------------------------------------
--------

>
>
> Juozas wrote:
>
> I haven't devoted much time to English phonetics and it seems

pretty

> inconsistent to me, at least in the assimilation field. For

example,

> the "g" sound in "dogpile" doesn't seem to turn into "k"

("dokpile"),

> whereas "it was just him" turns into "it wazh just him", so the

whole

> thing is quite confusing. In Lithuanian there are such pairs of
> consonants:
>
> b - p
> d - t
> g - k
> z - s
> zh - sh
> j (the one in "John") - ch
>
> The pairs mean, say, if "b" stands before any voiceless consonant

> becomes "p" and vice versa. Where I could find the list of

similar

> English consonant pairs? And it's not pure theory: if someone's
> ignoring consonant assimilation while speaking Lithuanian, it's
> clearly heard and unnatural (possibly same with English?).
>
> Juozas Rimas