Teenglish from England

Randy Alexander strangeguitars at GMAIL.COM
Fri Sep 18 08:00:42 UTC 2009

On Fri, Sep 18, 2009 at 3:40 PM, Nathan Sanders
<Nathan.Sanders at williams.edu> wrote:
>> When you know that when a sound sounds like something else it
>> actually IS the sound it sounds like then you'll understand.
> As M Covarrubias pointed out, you're conflating two very distinct
> things:
> (1) the physical reality: an articulation made in the speaker's vocal
> tract (consisting ofspecificed position and shape of the tongue, lips,
> velum, etc.), and the resulting sound wave that passes through that
> articulation (vibrating molecules with fundamental frequencies,
> harmonics, resonant frequencies, anti-formants, dampening, etc., as
> determined by the vocal articulation)
> (2) the auditory perception of that physical reality: the distortion
> made by the listener's ear and brain when they process a sound wave
> (there are a variety of relevant effects, such as perceptual averaging
> of neighboring resonant frequencies, non-linear transformation due to
> shape of the cochlea, the inherently logarithmic nature of perception
> as explained by Weber's Law, and even additional physical warping of
> the sound wave itself due to resonance and dampening within the ear
> canal)
> If you articulate [I] before [N], you will generate sound waves that,
> when measured by a computer, will come out with the expected
> frequencies for a nasalized [I] articulation, but when that sound wave
> is heard and interpreted by an actual human being with a human ear and
> human brain, it will often be perceived as [i] instead, because of how
> we transform the acoustic signal while processing it.
> Simply put, what we hear is not identical to what is said.

Of course Tom is going to say that what we hear should determine what
phoneme is represented, so before that happens I would like to stress
that it is only *some people* who perceive a nasalized [I] as [i].

Randy Alexander
Jilin City, China
My Manchu studies blog:

The American Dialect Society - http://www.americandialect.org

More information about the Ads-l mailing list