HF$IBBO:LIYM (?) -- (Parenthesis)

Ian Hutchesson mc2499 at mclink.it
Fri Sep 29 06:18:10 EDT 2000

At 12.28 28/09/00 -0500, Henry Churchyard wrote:
>>>>>> From: ben.crick at argonet.co.uk (Ben Crick)
>>>>>> There is no difference in pronunciation; how can you
>>>>>> differentiate in speech between "Sh" and "ShSh"?
>>>>> From: Henry Churchyard
>>>>> There certainly are languages that have a doubled "sh" sound in
>>>>> speech; Arabic and Japanese spring to mind...
>>>> From: Ian Hutchesson <mc2499 at mclink.it>
>>>> And in connected speech in English (ie spoken normally):
>>>>     Bush showed a lot.
>>>> There is a double $ going from Bush to showed. Try and say it with
>>>> only one and you end up with
>>>>     Bush owed a lot.
>>>> So, is there really no difference in pronunciation?
>>> From: ben.crick at argonet.co.uk (Ben Crick)
>>>  Well yes, there is. You put a glottal stop ('aleph or `ayin; or a
>>>  Shewa) between the last Sh of Bush and the first Sh of showed.
>>>  The same thing occurs with the often spoken "Bus stop", which
>>>  becomes "bus top" in lazy speech.  In Handel's Messiah, our choir
>>>  director insisted that we articulate "He shall reign for 'ever and
>>>  'ever" with glottal stops, not by singing it "he shall reign
>>>  forever endeavour".
>> From: Ian Hutchesson <mc2499 at mclink.it>
>> Definitely not! There is no glottal stop there in connected speech
>> -- that would be "unconnected" speech and quite unnatural!  There is
>> simply a double $, ie the $ is held from the end of Bush to the
>> start of showed.  In connected speech, although you don't put a
>> glottal stop between the two /s/, you can hear the difference
>> between "bus stop" and "bus top" because the /s/ is discretely
>> *longer* in the first. (This is not to contradict the idea that in
>> "lazy speech" only one /s/ is said.)
>Two points:
>1) English junctural phenomena are not a very good example here,
>because English does not have a distinctive "phonemic" consonant
>length contrast, so that the ears of native speakers of English are
>not attuned to hear length contrasts as being such (instead of as cues
>to possible word boundaries).  

The reason why I used a minimal situation (by necessity in the juncture
between words) is to show that while it is true "that the ears of native
speakers of English are not attuned to hear length contrasts" *within*
words, it is simply not true outside that situation. 

Answer me this: will you or will you not pronounce each of the following
(said normally, without insertions of glottal stops) in a different manner?

      1. Bush showed me a lot of money.
      2. Bush owed me a lot of money.

Will you or will you not hear that difference when someone else says it?

Time and again I have shown people arriving in Italy that they can already
recognize the difference between single and doubled consonants.

>Also, the junctural phenomena are
>inherently variable, subject to speech-style variation, etc. (i.e.
>are "gradient" rather than "categorical").  

The quibbling here is partially due to the fact that we are dealing with
fricatives (where the manifestation is one solely of length) and not
plosives which become even clearer. Try these

Bookkeeper   (ever heard "bookeeper"?)
Red door     (distinguishable from "red oar")

(And what about doubled affricates, eg "orange juice": ever heard "oran

There is another rather interesting phenomenon in English colloquial
connected speech which involves the assimilation of consonants: where two
plosives are consecutive, the first will often be assimilated, eg

Good boy   -> Goobboy
Hit parade -> Hipparade

Another case of doubled consonants in English. (It confuses the hell out of
foreign learners of English.)

Don't be shocked. Though people will swear blind that they don't say such
things, when recorded they are shown to. People will also rumble about this
not being good speech, but the rumbler will also be shown to do the same
thing, so one's notion of good speech doesn't reflect one's normal
production of language.

(Doubled plosives are normally manifested as the relevant consonant formed
then held for a moment before the plosion.)

>So it's much better to
>use as examples languages which have a phonemic -sh- vs. -shsh-
>contrast (such as the two I mentioned).  

The problem we are dealing with is that of native speakers of English who
don't have the conscious perception of doubled consonants. Therefore it is
a waste of time to talk about other languages until the notion has been
perceived. This is why I proposed a few simple functional examples in
English to see that it can be found there.

>Cross-linguistically, a
>distinctively geminated "sh" sound is not at all exotic or rare.
>2) "Bush showed" does _not_ contain a glottal stop between the two "sh"
>sounds in any normal type of speech.  A released glottal stop
>accompanied by aspiration (such as a glottal stop pronounced in
>conjunction with a following _sh_ sound would be) is a very distinctive
>"mini-cough" kind of sound, which is extremely recognizable once you've
>heard it and know what is.  And in combination, a released glottal stop
>preceding an "sh" sound would sound sort of like a "ch" (i.e. "tsh")
>affricate sound -- except with a mini-cough in place of the "t" ;-)
>However, "Bush owed" does contain a glottal-stop in a formal and
>consciously careful speech-style

Such accretions do not represent normal connected speech. When one becomes
conscious of what one is saying one stops speaking naturally.

>(with the glottal stop taken from the
>utterance-initial pronunciation of "owed", 

The perception here is a little askew: there is no implied glottal stop at
the beginning of owed; there is merely the insertion of a glottal stop at
the beginning of an utterance which would otherwise commence with a vowel,
so, while the 'utterance-initial pronunciation of "owed"' is preceded by a
glottal stop, there is no lurking glottal stop connected to "owed".

Such an insertion is totally unnatural.

>in order to push the "sh"
>unambiguously into the preceding syllable; this is the same
>pronunciation your choir teacher inisisted on).  In such a formal and
>consciously careful speech-style, "Bush showed" is actually pronounced
>_without_ such a glottal stop before the second vowel (in order to
>distinguish it from the formal careful pronunciation of "Bush owed").

This is all a little moot when the subject seems to be the perception of
doubled consonants rather than the avoidance of them through various
tricks. In connected speech -- the magic word is "connected", implying not
"formal careful" and unnatural -- no-one inserts a glottal stop in either
exemplar, yet the vast majority of speakers will make a clear distinction
between the two utterances. That distinction will amongst other things be
reflected in the lengthening of the $ to form what could be classed as a
double $ in languages in which such a distinction is recognized as meaningful.


More information about the b-hebrew mailing list