[KS] Romanization again

[John H. T. Harvey]

This is a multi-part message in MIME format.

------=_NextPart_000_0044_01BF5D58.D550A2C0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Dear List:=20

   I was interested by Prof. Lee Sangoak's findings on the preference =
for "k" over "g".

   Another reason for this preference, of course, perhaps too obvious to =
mention, is the ambiguity (in English and in several other languages) of =
"g," which leaves open the possibility that "Gim" is pronounced like =
"Jim."  This contrasts with "b" and "d," which are pretty unambiguous, =
as is "j" if you stick to English values.

    The current flurry of concern about the possible imposition of =
NAKL's proposed romanization scheme inspires me to address the topic =
again.  Perhaps it is not too late.  (A hearing was held on "eo," but =
none has so far been announced on "eu"!)
   =20
    Near the end of the last millennium I responded in a letter to the =
Korea Herald to a remark made by Prof. Shim Jae-kee in an interview in =
that paper to the effect that "we absolutely had to develop a system for =
use with the computer environment." I pointed out that, as far as I =
could determine, every Latin-alphabet writing system in the world =
(except, apparently, Bahasa Indonesian) uses either one or both of the =
two orthographic features he feels it necessary to eliminate, diacritics =
and the apostrophe, and that none seems to have been changed just =
because of the limitation of Internet domain names to plain letters, =
numerals, and the hyphen.

    Since I had rambled on for about as long as one is allowed to in a =
letter to a newspaper, I only hinted at my negative response to a quite =
different comment of Prof. Shim's in that interview, namely that the =
proposed new system presents "the true linguistic integrity and identity =
of the Korean language."  This is a point frequently made by proponents =
of the NAKL proposal, as it was, for example, by Prof. Kim Se-jung at =
the latest hearing.  I would like to take it up now even though Prof. =
Shim told foreigners that they are "kindly advised to adapt and learn" =
(I have teased him about this remark, but it may have sounded a bit =
friendlier in the original Korean, of course), and even though this =
second topic is one on which foreigners might well be assumed to have =
even less right to comment.

    First, it is often pointed out that the distinctions the current =
official system makes between voiced and voiceless plosives are =
allophonic -- that is, not phonemic, below the attention level and even =
below the perception level of native speakers.  This is true.   =
(Interestingly, however, I am told that NAKL's own guide to the =
"kanafication" of Korean, Kugo^u^i Kana Muncha P'yogibo^p, advises the =
use of the kana voicing diacritic ("nigori") to indicate the allophonic =
voicing of the plain plosives.) It is also true of the distinction =
between "s" and "sh."  However, it is just as true of the distinction =
between "r" and "l," which the NAKL proposal *does* distinguish.  I have =
not come across any justification for this exception to what we are told =
is a principle of vital importance, although it has occurred to me that =
perhaps the Western stereotype of Orientals (including, =
counterfactually, Mandarin speakers) as not being able to make any such =
distinction could be behind it.  I note that the otherwise pure =
transliteration used in the discussion of Han-gu^l in The Unicode =
Standard, Version 2.0, also insists on this one exception, rendering the =
name of the phoneme in question "rieul."=20

    Second, how can it be that "the true linguistic integrity and =
identity of the Korean
language" is captured by a system that writes initial aspirated stops =
and final unreleased stops with the same letters ("k, t, p," as in =
"Koria (Hereoldeu)" and "Hanguk," for example)?  Surely it is absolutely =
clear to speakers of Korean that it is the initial *unaspirated* stops =
that are the same phonemes as the final unreleased stops.  (This despite =
the fact that aspirated stops written in final position are =
morphophonemically reduced to unreleased stops when pronounced finally.) =
 This measure seems to be a complete violation of "the true linguistic =
integrity and identify of the Korean language."  It can only be =
understood as another concession to foreigners, doubtless motivated by =
all the claims in the last two go-rounds that final voiced consonant =
letters, as in "Bag Seonsaeng," if pronounced voiced, sound "dumb."=20

    Third, a major reason given for a new romanization system is the =
difficulty Koreans have in following the rules of the current official =
system (which reflect rules they apply automatically when they speak but =
which they find hard to raise to the level of consciousness).  This is =
certainly a valid consideration, and a strong argument for having at =
least a supplementary transliteration system.  But consider the NAKL =
distinction between initial and final stops which we just discussed.  =
Let's say that a native speaker learns to write "guk," the lax initial =
with a voiced letter and the equivalent final with a voiceless letter =
used initially for an aspirate.  Now he wants to write 'national =
language,' which is "guk + eo."  But he can't just write the two =
syllables together.  That would yield "gukeo," with the "k" probably =
interpreted as an initial and therefore as an aspirate.  So he has to =
change the "k" back to the "g" he probably wanted to use in the first =
place:  "gugeo."  Judging from the example "Seokkuram," also, "k + g" =
equals "kk," which does not equal "k + k"! Although "k + j" equals "kj" =
in "Paekje," not "kch"!  Is all this any easier than using the voiced =
consonant letters between other voiced letters (such as vowels, nasals, =
and "l/r"), as in the current official system?  I suspect that this =
"liaison" rule would make even less sense to the average native =
romanizer.  At any rate, it is hard to see any gain worth redoing all =
the road signs and guidebooks.  (Unless romanization revision is a =
make-work project!)

    Fourth, somewhat less seriously, it has been said to be somehow =
against the spirit of Han-gu^l to use apostrophes to mark the aspirated =
plosives.  But isn't that remarkably close to what King Sejong himself =
did?  He added strokes to the letters of "k" and "t" and something very =
much like an apostrophe on top of the letter for "ch."  In the case of =
the aspirated version of "p" he did not actually add anything, but this =
has been explained reasonably as avoiding having a single letter with as =
many as five strokes.  The apostrophe is used for aspiration in the =
Wade-Giles system for Chinese and for the glottal stop in other =
alphabets and romanization systems, as well as for deletions in several =
writing systems.

    I would say that the case that the proposed system better reflects =
the phonological structure of Korean and/or would be easier for Koreans =
to use is far from proven, and certainly cannot carry the burden of =
proof needed to justify inflicting yet another system of Korean =
romanization on Koreans and others who want to write about Korea in the =
Latin alphabet.=20

    It might be wise to remember that the 1959 system (with its infamous =
"Dogribmun" and "Geobugseon") still lurks in many minds, not to mention =
names, while foreign scholars and agencies still almost invariably use =
either unmodified McCune-Reischauer (admittedly almost indistinguishable =
from the current official system) or the Yale-Martin transliteration. =20

    On top of the old saying "If it ain't broke, don't fix it!" we =
should heed add the even older saying "Above all, do no harm!"

John H. T. Harvey
jharvey at nuri.net


------=_NextPart_000_0044_01BF5D58.D550A2C0
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><FONT size=3D3><FONT size=3D3>Dear List:</FONT> </FONT></DIV>
<DIV><FONT size=3D3></FONT> </DIV>
<DIV><FONT size=3D3>   I was interested by Prof. Lee Sangoak's =
findings=20
on the preference for "k" over "g".</FONT></DIV>
<DIV> </DIV>
<DIV>   Another reason for this preference, of course, =
perhaps=20
too obvious to mention, is the ambiguity (in English and in several =
other=20
languages) of "g," which leaves open the possibility that "Gim" is =
pronounced=20
like "Jim."  This contrasts with "b" and "d," which are pretty =

unambiguous, as is "j" if you stick to English values.</DIV>
<DIV> </DIV>
<DIV>    The current flurry of concern about the possible =

imposition of NAKL's proposed romanization scheme inspires me to address =
the=20
topic again.  Perhaps it is not too late.  (A hearing was held =
on=20
"eo," but none has so far been announced on "eu"!)</DIV>
<DIV>    </DIV>
<DIV>    Near the end of the last millennium I responded =
in a=20
letter to the Korea Herald to a remark made by Prof. Shim Jae-kee in an=20
interview in that paper to the effect that "we absolutely had to develop =
a=20
system for use with the computer environment." I pointed out that, as =
far as I=20
could determine, every Latin-alphabet writing system in the world =
(except,=20
apparently, Bahasa Indonesian) uses either one or both of the two =
orthographic=20
features he feels it necessary to eliminate, diacritics and =
the=20
apostrophe, and that none seems to have been changed just because =
of the=20
limitation of Internet domain names to plain letters, numerals, and =
the=20
hyphen.</DIV>
<DIV> </DIV>
<DIV>    Since I had rambled on for about as long as one =
is=20
allowed to in a letter to a newspaper, I only hinted at my negative =
response to=20
a quite different comment of Prof. Shim's in that interview, namely that =
the=20
proposed new system presents "the true linguistic integrity and identity =
of the=20
Korean language."  This is a point frequently made by proponents of =
the=20
NAKL proposal, as it was, for example, by Prof. Kim Se-jung at the =
latest=20
hearing.  I would like to take it up now even though =
Prof. Shim=20
told foreigners that they are "kindly advised to adapt and learn" (I =
have teased=20
him about this remark, but it may have sounded a bit friendlier in the =
original=20
Korean, of course), and even though this second topic is one on which =
foreigners=20
might well be assumed to have even less right to comment.</DIV>
<DIV> </DIV>
<DIV>    First, it is often pointed out that the =
distinctions=20
the current official system makes between voiced and voiceless =
plosives are=20
allophonic -- that is, not phonemic, below the attention level and even =
below=20
the perception level of native speakers.  This is=20
true.   (Interestingly, however, I am told that NAKL's own =
guide to=20
the "kanafication" of Korean, Kugo^u^i Kana Muncha P'yogibo^p, advises =
the use=20
of the kana voicing diacritic ("nigori") to indicate the allophonic =
voicing of=20
the plain plosives.) It is also true of the distinction between "s" and=20
"sh."  However, it is just as true of the distinction between "r" =
and "l,"=20
which the NAKL proposal *does* distinguish.  I have not come across =
any=20
justification for this exception to what we are told is a principle of =
vital=20
importance, although it has occurred to me that perhaps the Western =
stereotype=20
of Orientals (including, counterfactually, Mandarin speakers) as not =
being able=20
to make any such distinction could be behind it.  I note=20
that the otherwise pure transliteration used in the discussion of =
Han-gu^l=20
in The Unicode Standard, Version 2.0, also insists on this one =
exception,=20
rendering the name of the phoneme in question "rieul." </DIV>
<DIV> </DIV>
<DIV>    Second, how can it be that "the true linguistic=20
integrity and identity of the Korean<BR>language" is captured by a =
system that=20
writes initial aspirated stops and final unreleased stops with the =
same=20
letters ("k, t, p," as in "Koria (Hereoldeu)" and "Hanguk," for =
example)? =20
Surely it is absolutely clear to speakers of Korean that it is the =
initial=20
*unaspirated* stops that are the same phonemes as the final unreleased=20
stops.  (This despite the fact that aspirated stops written in =
final=20
position are morphophonemically reduced to unreleased stops when =
pronounced=20
finally.)  This measure seems to be a complete violation =
of "the true=20
linguistic integrity and identify of the Korean language."  It can =
only be=20
understood as another concession to foreigners, doubtless motivated by =
all the=20
claims in the last two go-rounds that final voiced consonant letters, as =
in "Bag=20
Seonsaeng," if pronounced voiced, sound "dumb." </DIV>
<DIV> </DIV>
<DIV>    Third, a major reason given for a new =
romanization=20
system is the difficulty Koreans have in following the rules of the =
current=20
official system (which reflect rules they apply automatically when =
they=20
speak but which they find hard to raise to the level of=20
consciousness).  This is certainly a valid consideration, and a =
strong=20
argument for having at least a supplementary transliteration =
system.  But=20
consider the NAKL distinction between initial and final stops which we =
just=20
discussed.  Let's say that a native speaker learns to write "guk," =
the lax=20
initial with a voiced letter and the equivalent final with a voiceless =
letter=20
used initially for an aspirate.  Now he wants to write 'national =
language,'=20
which is "guk + eo."  But he can't just write the two syllables =
together. =20
That would yield "gukeo," with the "k" probably interpreted as an =
initial and=20
therefore as an aspirate.  So he has to change the "k" back to the =
"g" he=20
probably wanted to use in the first place:  "gugeo."  Judging =
from the=20
example "Seokkuram," also, "k + g" equals "kk," which does not equal "k =
+ k"!=20
Although "k + j" equals "kj" in "Paekje," not "kch"!  Is all this any =
easier=20
than using the voiced consonant letters between other voiced letters =
(such as=20
vowels, nasals, and "l/r"), as in the current official system?  I =
suspect=20
that this "liaison" rule would make even less sense to the average=20
native romanizer.  At any rate, it is hard to see any gain=20
worth redoing all the road signs and guidebooks.  (Unless =

romanization revision is a make-work project!)</DIV>
<DIV> </DIV>
<DIV>    Fourth, somewhat less seriously, it has been =
said to be=20
somehow against the spirit of Han-gu^l to use apostrophes to mark the =
aspirated=20
plosives.  But isn't that remarkably close to what King Sejong =
himself=20
did?  He added strokes to the letters of "k" and "t" and something =
very=20
much like an apostrophe on top of the letter for "ch."  In the case =
of the=20
aspirated version of "p" he did not actually add anything, but this has =
been=20
explained reasonably as avoiding having a single letter with as many as =
five=20
strokes.  The apostrophe is used for aspiration in =
the Wade-Giles=20
system for Chinese and for the glottal stop in other alphabets and =
romanization=20
systems, as well as for deletions in several writing systems.</DIV>
<DIV> </DIV>
<DIV>    I would say that the case that the proposed =
system=20
better reflects the phonological structure of Korean and/or would be =
easier for=20
Koreans to use is far from proven, and certainly cannot carry the burden =
of=20
proof needed to justify inflicting yet another system of Korean =
romanization on=20
Koreans and others who want to write about Korea in the Latin alphabet. =
</DIV>
<DIV> </DIV>
<DIV>    It might be wise to remember that the 1959 =
system (with=20
its infamous "Dogribmun" and "Geobugseon") still lurks in many minds, =
not to=20
mention names, while foreign scholars and agencies still almost =
invariably use=20
either unmodified McCune-Reischauer (admittedly almost indistinguishable =
from=20
the current official system) or the Yale-Martin transliteration.  =
</DIV>
<DIV> </DIV>
<DIV>    On top of the old saying "If it ain't broke, =
don't fix=20
it!" we should heed add the even older saying "Above all, do =
no=20
harm!"</DIV>
<DIV> </DIV>
<DIV>John H. T. Harvey</DIV>
<DIV><A href=3D"mailto:jharvey at nuri.net">jharvey at nuri.net</A></DIV>
<DIV> </DIV></BODY></HTML>

------=_NextPart_000_0044_01BF5D58.D550A2C0--



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