Re: Prosody-only utterance

From: Douglas G. Kilday (fufluns_at_chorus.net)
Date: 01/09/05 

Date: Sun, 9 Jan 2005 16:44:24 -0000

"Bobby D. Bryant" <bdbryant@mail.utexas.edu> wrote ...
>
> Among anglophones, especially children, the prase "I don't know" is
> sometimes uttered in a very reduced form, without any consonantal
> segments, and the vowels (if vowels they are) are a neutral sound
> without any relation to the vowels in the ordinary phrase; but the
> pitch, stress, and timing of the ordinary phrase are retained. (I
> think it can also be hummed rather than vocalized.)

I use the "hummed", i.e. fully nasalized form, when responding to questions
with my mouth full. It amounts to [?m:] with pitch contour roughly [2312].
The "information" is carried by the pitch contour. This form is very useful
when one's girlfriend has the habit of asking questions while one is eating.

> Questions: Do you consider this speech? Can you think of any other
> phrases that get the same treatment, in English or other languages?
> Is there a name for the phenomonen? Has anyone investigated what
> rules, if any, govern its application? Do we know anything about the
> history of this treatment for "I don't know" in English?

I consider this speech. So are whistled forms of certain languages, in my
opinion. But I draw the line when it comes to drummed forms of certain
languages, stereotyped tapping for clubhouse entry, etc. which are loosely
equivalent to Morse Code telegraphy and only _based_ on speech.

I once observed a professor produce a fully nasalized form of "Hey!" when a
student was about to put a completed test in the wrong place. The professor
was chewing tobacco at the time, and open oral utterance would have created
a mess. One might expect communities of tobacco-chewers to have a larger
inventory of nasalized utterances than ordinary speakers.

> Somewhat similarly, though perhaps not related to the above, we get
> the bisyllabic forms of 'yes' and 'no' in English, which I will
> venture to transcribe as -
>
> 'yes' : /?^h^/ - an iamb with rising pitch
> 'no' : /h^?^/ - a trochee with falling pitch
>
> These can also be hummed, I think with nazalization on the first
> syllable of the 'no' form, something like -
>
> 'no' : /hn?m/ - where /n/ and /m/ are syllabic resonants
>
> The hummed 'yes' form seems to be subject to prolongation beyond the
> normal length of syllables, and variations in the pitch and stress as
> well.

I use the nasalized forms often, and regard them as [?m?m] 'no' and [?mhm]
'yes'. Since I regard them as part of English speech, I consider the pair
to prove that English has a phonemic glottal stop, as /?/ and /h/ provide a
minimal-pair contrast. Variations in pitch and stress convey emotional
information which may be equivalent to spoken information. E.g.:

[Laxative commercial circa 1973]
Wife: "Is it ... irregularity?"
Husband: "Mmhmm! ... But I'm _not_ taking any harsh laxatives!"
(Here the first syllable of the affirmation was prolonged at uniform pitch;
the second rose rapidly and was abruptly cut off, apparently by a glottal
stop. The pitch contour conveyed a somewhat disgusted resignation toward
the fact of the speaker's constipation.)

[Telephone conversation circa 1986]
Former squeeze: "If you come here, we can do something."
DGK: "Mmhmm!"
FS: "Not _that_!"
(Here the second syllable of the affirmation abruptly rose and fell in
pitch, with intensity trailing off like a "wolf whistle", suggesting amorous
intent. The demonstrative referred to a specific potential act implied by
the contour, so it might be argued that pitch/intensity contours sometimes
convey _lexical_ information.)

> The vocalized forms seem to be 'designed' to maximize contrast when
> working with a very limited set of segments. (They might be seen as a
> pair with a reversal of the order of the syllables, dragging the pitch
> and stress along with the reversal, though that "explanation" may just
> be the natural fallout of maximized contrast when using just a few
> binary features.)
>
> And I have similar questions for these too: Do you consider them
> words? Can we derive them from anything? Is there a name for this
> phenomonen? Is it related to the "I don't know" form? Are there
> similar phenomena (or other manifestations of the same thing) in other
> languages?

I consider them words, and like the "I don't know" form, they are very
useful for answering questions while eating. I don't know how they
originated, or how old they are. Possibly the aspiration in the affirmative
form represents the approximant in "yea", and possibly the negative form was
originally a reduplication of the old negative particle "ne", but I am only
guessing. Somebody here ought to know.

Relevant Pages

  • ESPS method for pitch detection
    ... as input and gives pitch contour (frame level pitch values) ... as output by using ESPS method of pitch extraction. ...
    (comp.soft-sys.matlab)
  • Re: Another pronounciation question - hotaru ga ike
    ... have difficulty hearing pitch properly. ... own language tends to pronounce stressed syllables at a higher pitch, ... we hear non-existent stress depending on the pitch pattern of Japanese ...
    (sci.lang.japan)
  • Re: why cant the BBC
    ... Moreover, the Hz's of formants are the pitch of any actual tone, but ... at exactly the same pitch and the listener can still reliably tell ... speaker to make contrastive distinction between vowels. ...
    (sci.lang)
  • Re: Another pronounciation question - hotaru ga ike
    ... as a speaker of a stress-accented-language, Japanese people don't care. ... And I think you are partly right; my ear does generally tend to register high pitch as carrying a stronger stress, and I do sometimes have difficulty hearing pitch properly. ... I think this is common for native English speakers; our own language tends to pronounce stressed syllables at a higher pitch, so we hear non-existent stress depending on the pitch pattern of Japanese words. ...
    (sci.lang.japan)
  • =?ISO-8859-1?Q?Re=3A_Redefining_=5Ba=5D_and_=5B=E6=5D=3F_or_using_a_better_vowel?= =?ISO-
    ... F0 (pitch) between the following two lines in a hymn might have the ... F1 and F2 are resonance frequencies, based on the shape of the mouth ... Of course, there will be minor variations in the formants, because ... (I even have a tape of a singer singing the five vowels at frequencies ...
    (sci.lang)