[KS] NAKL Questionnaire (2)

Wed Oct 13 10:07:37 EDT 1999

This is a multi-part message in MIME format.

------=_NextPart_000_002A_01BF15CF.BD8B4DA0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

=20
The NAKL Romanization Questionnaire=20

2. Methodology=20

I have several reservations about the methodology of the NAKL =
Romanization questionnaire.=20

First, languages are systems, and so surely are best represented by =
systematic orthographies. But the questionnaire does not ask respondents =
to choose between systems. It asks them about one minor point after =
another.=20

So, to take an admittedly extreme example, it would be quite possible =
for a respondent in one place to endorse the apostrophe to mark =
aspiration and in another to endorse it to replace the breve accent mark =
of the current system. Would that mean that he was really voting for =
Inch'o'n and the like? Probably not. It's hard to imagine anyone really =
wanting this. But such votes are possible. How will they be tallied?=20

When it comes to all the options presented for representing the vowels, =
I suspect that even a specialist could get confused.=20

Surely it would be better to ask respondents to choose between carefully =
thought out, coherent systems=97-say the current system, the (almost =
identical) original McCune-Reischauer system, the old MOE system, =
Martin's Yale system, new phonetic systems A and B, new transliteration =
systems A and B, and so on.=20

Second, a number of questions present themselves about the polling =
process.=20

For example, the appropriateness of polling in the first place. I rather =
doubt that Sejong polled even the Hall of Worthies. Certainly, if he had =
polled the literati more widely, we wouldn't have Han'g=FBl today!=20

Or the representativeness of the group polled. Surely it will =
disproportionately represent native speakers of Korean. So it would not =
be surprising if ease of Romanization, of great practical importance to =
Koreans, who do most of the Romanizing, outweighs what might be called =
"phonetic transparency," of equally great practical importance to =
foreigners, who are the major consumers of Romanization. (China's Pinyin =
system is very simple for speakers of Mandarin, but it takes foreigners =
about as long to learn what c, q, and x represent, or, for that matter, =
many of the other consonants and most of the vowels, as it would for =
them to learn Han'g=FBl! The Chinese can get away with it because they =
are one-fifth of the world.)=20

Or the other differences in the group in backgrounds and interests. Some =
of the activists in the re-Romanization movement seem to be particularly =
concerned with using computer systems which favor the English alphabet, =
particularly for worldwide communication with specialists in things =
Korean. Will the average respondent have any appreciation of these =
matters? Will he, for instance, be aware of how fast computers are =
improving in handling language complexities? Will he have any =
appreciation of the almost universal use of the McCune-Reischauer system =
by Koreanists or of the Yale system by linguists?=20

To veer away from the methodology of the questionnaire, I do not, =
myself, feel that present computer limitations are an argument for =
re-opening the Romanization question. Quite the reverse, in fact. As =
suggested above, electronic language processing is improving too rapidly =
and surprisingly at this cusp of centuries and millennia for it to be a =
good time to make decisions which, if we are to judge from the last two =
Romanization systems, will be with us for decades. (As a matter of fact, =
the voiced letters and the eo and eu of the pre-1984 system still lurk =
in the Korean subconscious!)=20

If you consider the miracle already accomplished that allows us to type =
one Han'g=FBl letter after another and leave it to the computer to =
arrange them properly into syllables (before electronic typewriters, the =
keyboard had to have positional variants of each consonant and vowel), =
the feasibility is obvious of a program that automatically transforms =
Han'g=FBl into either a transliteration or a reasonably phonetic =
representation like the current system. It could even keep the =
transliteration as an underlying representation when displaying the =
phonetic representation, allowing the Han'g=FBl to be recovered. So it =
might be quite a waste of time, effort, money, and peace of mind to =
change the Romanization system now without having a very clear idea of =
what the next few years might bring.

John H. T. Harvey

jharvey at nuri.net

------=_NextPart_000_002A_01BF15CF.BD8B4DA0
Content-Type: text/html;
	charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Dwindows-1252" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV> 
The NAKL Romanization Questionnaire=20
2. Methodology=20
I have several reservations about the methodology of =
the NAKL=20
Romanization questionnaire.=20
First, languages are systems, and so surely are best=20
represented by systematic orthographies. But the questionnaire does not =
ask=20
respondents to choose between systems. It asks them about one minor =
point after=20
another.=20
So, to take an admittedly extreme example, it would =
be quite=20
possible for a respondent in one place to endorse the apostrophe to mark =

aspiration and in another to endorse it to replace the breve accent mark =
of the=20
current system. Would that mean that he was really voting for =
Inch'o'n=20
and the like? Probably not. It's hard to imagine anyone really =
wanting this.=20
But such votes are possible. How will they be tallied?=20
When it comes to all the options presented for =
representing the=20
vowels, I suspect that even a specialist could get confused.=20
Surely it would be better to ask respondents to =
choose between=20
carefully thought out, coherent systems=97-say the current system, the =
(almost=20
identical) original McCune-Reischauer system, the old MOE system, =
Martin's Yale=20
system, new phonetic systems A and B, new transliteration systems A and =
B, and=20
so on.=20
Second, a number of questions present themselves =
about the=20
polling process.=20
For example, the appropriateness of polling in the =
first place.=20
I rather doubt that Sejong polled even the Hall of Worthies. Certainly, =
if he=20
had polled the literati more widely, we wouldn't have Han'g=FBl =
today!=20
Or the representativeness of the group polled. Surely =
it will=20
disproportionately represent native speakers of Korean. So it would not =
be=20
surprising if ease of Romanization, of great practical importance to =
Koreans,=20
who do most of the Romanizing, outweighs what might be called "phonetic=20
transparency," of equally great practical importance to foreigners, who =
are the=20
major consumers of Romanization. (China's Pinyin system is very simple =
for=20
speakers of Mandarin, but it takes foreigners about as long to learn =
what=20
c, q, and x represent, or, for that matter, =
many of=20
the other consonants and most of the vowels, as it would for them to =
learn=20
Han'g=FBl! The Chinese can get away with it because they are one-fifth =
of the=20
world.)=20
Or the other differences in the group in backgrounds =
and=20
interests. Some of the activists in the re-Romanization movement seem to =
be=20
particularly concerned with using computer systems which favor the =
English=20
alphabet, particularly for worldwide communication with specialists in =
things=20
Korean. Will the average respondent have any appreciation of these =
matters? Will=20
he, for instance, be aware of how fast computers are improving in =
handling=20
language complexities? Will he have any appreciation of the almost =
universal use=20
of the McCune-Reischauer system by Koreanists or of the Yale system by=20
linguists?=20
To veer away from the methodology of the =
questionnaire, I do=20
not, myself, feel that present computer limitations are an argument for=20
re-opening the Romanization question. Quite the reverse, in fact. As =
suggested=20
above, electronic language processing is improving too rapidly and =
surprisingly=20
at this cusp of centuries and millennia for it to be a good time to make =

decisions which, if we are to judge from the last two Romanization =
systems, will=20
be with us for decades. (As a matter of fact, the voiced letters and the =
eo=20
and eu of the pre-1984 system still lurk in the Korean=20
subconscious!)=20
If you consider the miracle already accomplished that =
allows us=20
to type one Han'g=FBl letter after another and leave it to the computer =
to arrange=20
them properly into syllables (before electronic typewriters, the =
keyboard had to=20
have positional variants of each consonant and vowel), the feasibility =
is=20
obvious of a program that automatically transforms Han'g=FBl into either =
a=20
transliteration or a reasonably phonetic representation like the current =
system.=20
It could even keep the transliteration as an underlying representation =
when=20
displaying the phonetic representation, allowing the Han'g=FBl to be =
recovered. So=20
it might be quite a waste of time, effort, money, and peace of mind to =
change=20
the Romanization system now without having a very clear idea of what the =
next=20
few years might bring.
John H. T. Harvey
<A=20
href=3D"mailto:jharvey at nuri.net">jharvey at nuri.net</A></DIV></B=
ODY></HTML>

------=_NextPart_000_002A_01BF15CF.BD8B4DA0--

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%