Roanoke Times
Copyright (c) 1995, Landmark Communications, Inc.
DATE: THURSDAY, March 2, 1995 TAG: 9503040039
SECTION: EXTRA PAGE: 6 EDITION: METRO
SOURCE: JOEL ACHENBACH
DATELINE: LENGTH: Medium
A: Let's say you work for the Why staff and you want to call up an expert on Spam. You look in the Why Rolodex under S, get the number, call the Spam HQ, and explain that you need some information about the pink canned meat product. But the dumb bunny on the other end of the line says, ``Stan?''
And so you have to go through the whole alphabetic deconstruction. ``No,'' you say, ``Spam - `s' as in sphincter, `p' as in pfennig, `a' as in aesthetic, `m' as in mnemonic.'' Communicating clearly by phone is an incredible hassle.
According to AT&T Bell Laboratories, the two letters hardest to distinguish over the phone are ``f'' and ``s.'' This sounds odd, since in regular face-to-face conversation there's no difficulty at all in telling the difference between the sounds. There's an obvious difference between ``fly'' and ``sly'' or between ``fizzle'' and ``sizzle.''
What you don't realize when you are talking to someone face to face is that you are unconsciously lip-reading. When you make an ``f'' sound, air rushes through a small opening between your teeth and your lips. With an ``s,'' air rushes through a small opening between the tip of your tongue and the back side of your front teeth. Since in both cases you have air rushing through a small passage, the sounds are similar.
Moreover - and here's the crucial part of the answer - the ``f'' and ``s'' sounds are emitted at high frequencies, between 3000 and 6000 hertz. But the telephone system only handles frequencies between 200 to 4000 hertz. Thus most of the ``information'' in the ``f'' and ``s'' sounds, all the nuance above 4000 hertz, is lopped off by the telephone. The phone captures just enough of the difference to allow you to usually tell an ``s'' from an ``f,'' but people still make a mistake about 10 percent of time, says David Roe, head of the applied speech research department at Bell Labs.
After the ``f'' and the ``s,'' the two sounds hardest to distinguish are ``m'' and ``n.'' For our money they don't even deserve to be separate letters. (We're told that the ``m'' and ``n'' sounds come out of the nose. But we did a test. We discovered that we use the mouth for those. Maybe we're just doing it wrong.)
The third most ambiguous pair is ``p'' and ``t,'' fourth is ``b'' and ``d,'' and fifth is ``g'' and ``t.''
Now you might ask, why doesn't the phone company invent a telephone with broader bandwidth, so that it could pick up the subtleties above 4000 hertz? There are, in fact, experimental telephones that digitize the signal directly at the handset, and have greater fidelity. But even in the age of digital technology and fiber-optic lines, the average telephone is still connected to old-fashioned twisted-pair lines.
There's also the fact that phone calls are bilateral. No one wants to spend a lot of money for some fancy new broad-bandwidth telephone if no one else has a phone that can receive the improved signal. Besides which, the old standard is perfectly adequate for delivering most of the range of human intelligibility.
So stop complaining. (Stop, we say. That's ``s'' as in schlemiel, ``t'' as in - Washington Post Writers Group
by CNB