[KS] NAKL's New Romanization Proposal

John H. T. Harvey jharvey at nuri.net
Mon Nov 29 11:00:37 EST 1999


This is a multi-part message in MIME format.

------=_NextPart_000_0037_01BF3ACE.506C1DC0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

=20
SOME NOTES ON THE ROMANIZATION PROPOSAL PUT FORWARD BY THE NATIONAL =
ACADEMY OF THE KOREAN LANGUAGE AT A HEARING ON NOVEMBER 19, 1999

=20

John H. T. Harvey

=20

     I attended the hearing at the kind invitation to the list, on =
behalf of NAKL, by Prof. Lee Sang Oak.  I have not yet had time to study =
the materials handed out at the hearing as thoroughly as I should, but =
as that could take some time, and as I was the only foreigner who =
attended and received them, I thought I should report my first =
impressions to the list and to NAKL. I'm sure the materials will be =
available from NAKL.

=20

[Some Notes On These Notes: =20

Trying to avoid every conceivable problem that could be caused by =
different operating systems, browsers, and word processors, I have used =
good old CAPITALS (instead of the usual italics) for cited romanized =
Korean words and letters, using a prefixed asterisk (*) to indicate =
romanizations proposed by NAKL. I have written a following caret (^) =
instead of the current breve ("half moon") diacritic for the obscure =
vowels in INCH'O^N and KU^MGANGSAN. I have also tried to avoid any =
formatting that might not get through, however desperately Word 2000 =
insisted on it, except for indenting paragraphs, which I don't think can =
do much harm.]

=20

"The Romanization of Korean" (Revised Proposal)

p. 1

Chapter 1. The Basic Principles of Romanization

         1. The romanization of Korean shall follow the standard =
pronunciation of Korean.

         2. No symbols shall be used other than the letters of the Roman =
alphabet.

         3. The principle shall be "one sound =97 one letter."

=20

JHTH NOTES:

         Principle 1, to many, may seem almost too self-evident to need =
stating. It should presumably be taken as referring to the adoption of =
phonemic (or phonetic) representation over the alternative of =
transliteration (i.e., retaining the morphophonemic nature of Han-gu^l =
orthography) as in the '59 MOE system and the '97 proposal of this same =
body.=20

         Principle 2 uses a very strict interpretation of "letters of =
the Roman alphabet," one that would rule out all the accented letters of =
French, Spanish, German, and most other Western European languages that =
use the Roman alphabet (i.e., it rejects the breve accent mark of the =
present system) as well as ruling out the apostrophe so widely used in =
English orthography (although admittedly not for phonetic purposes).

         Principle 3 seems not to be applied very strictly, at least in =
the sense that this revised proposal adds two digraphs for single vowels =
(*EO and *EU, as in the '59 system) to the two (AE and OE) of the =
current ('84) system, while keeping the consonant digraphs (CH, NG, and =
double consonants). So perhaps this principle refers instead to the fact =
that no distinction is made between the voiceless allophones (K, T, P, =
CH) and the voiced allophones (G, D, B, and J) of the lax stops and =
affricate, nor between the unpalatalized (S) and palatalized (SH) of the =
lax fricative. Note, however, that a distinction is still made between =
the flap (R) and the lateral (L) values of the liquid. All these =
differences are, of course, below the phonemic level =97 that is, the =
level of phonetic awareness =97 for native speakers of Korean, who =
therefore have difficulty reflecting them in the romanization.  One can =
only assume that NAKL was afraid that the R/L contrast was a much more =
sensitive matter than voicing and palatalization, to the extent that =
Koreans should have to learn to make the distinction.

=20

pp. 1-6

Chapter 2. A Glance at the Romanization

=20

JHTH NOTES:

          In the vowels, the only change from the current system is the =
elimination of the breve accent mark by reversion to *EO and *EU for O^ =
and U^, changes apparently motivated by antipathy to the breve (or "half =
moon") accent mark.  Anticipating the revival of complaints about the =
failure of these spellings to suggest anything like the sounds intended, =
NAKL elsewhere commits itself to a vigorous campaign to educate =
foreigners.  This will give them several hundred thousand students a =
year, I believe, most of them fleeting targets.

         In the diphthongs, the only surprises are *WO, not *WEO, for =
what should be WO^ but is actually WO in the current system and, =
correspondingly, *UI, not *EUI, for U^I.  Both simplifications ignore =
the foreign user. It may be obvious to Korean speakers that *WO must be =
*W+EO, since there is no W+O in the language, but this is hardly the =
case for foreigners. In my own dialect, the former is rather like the =
diphthong in 'work' while the latter would be rather like that in =
'walk,' both equally possible. (As noted, the current system has the =
same misguided economy.) Similarly, there is no way for a less than =
fluent foreigner to know when *UI in the proposed system stands for =
*EU+I and when it stands for *U+I, as in *UIDONG, unless the latter is =
hyphenated, *U-IDONG. *UI does, I grant, keep *YEOUIDO down to four =
vowel letters in a row! By the way, I was delighted to find HU^I =
romanized *HI in the example *KWANGHIMUN. The present romanization here =
resorts to transliteration rather than guiding the foreigner to the =
correct pronunciation. The same could be said of the Y in cases like =
*TOEGYERO, I believe. At least, I don't hear it.=20

         The proposed consonant system seems to focus on eliminating the =
present system's apostrophe for aspiration without using final voiced =
letters, which have been criticized as sounding ugly. So Mr. PAK would =
be not Mr. BAG but Mr. *BAK. This has a lot to be said for it (which in =
fact was said for it to the Korean Studies list by Gary Rector a couple =
of years back).  What can't be said for it is that it is an =
instantiation of the "one sound =97 one letter" principle, since it =
writes an aspirated consonant the same way as the equivalent unreleased =
final stop, even though the two are phonemically distinct in the =
language. This ignoring of the phonetic intuition of Koreans could well =
introduce for them a difficulty very similar to the one removed by not =
distinguishing voiced and voiceless plosives.=20

         I can't find an example in the proposal of a syllable-final =
stop followed by a vowel or glide, but I presume that *BUK+AK would have =
to become *BUGAK, by liaison.  Otherwise you would have BUKAK, where the =
first K would indicate an aspirate (although the second K, of course, =
would be unreleased). This could only add to the difficulties of the =
native speaker romanizer.  (Actually, I did find *YUN SUKYEONG, also =
written *YUN SUK-YEONG and *YUN SUK YONG, but can't believe that this is =
intended seriously. What would be the point of writing *SUK+YEONG =
differently from *SU+GYEONG, when they are pronounced the same, while =
writing *SUK+YEONG the same as *SU+KYEONG, when they would be pronounced =
differently?)

         Reinforced J is to be written *JJ, a cosmetic change that =
weakens the phonetic "meaning" of the voiced letters.

         The reinforced consonants (the others written by doubled =
voiceless letters, as at present) present another problem for the Korean =
user of the system.  The rules of the proposed system would write a =
syllable-final stop and a following syllable-initial lax stop as a =
voiceless letter followed by a voiced letter, as in the example *GAPGOT. =
However, if the consonants are homorganic, that combination is =
phonetically indistinguishable from the reinforced consonant, which is =
written with a doubled voiceless letter. Compare their examples =
*SOKGURAM and *JJILLEKKOT.  Gary Rector nicely got around this problem =
by writing the reinforced consonants with voiced/voiceless pairs:  KG, =
TD, PB, and TJ, a choice for which he also gave an interesting phonetic =
justification. This regularity might make things easier for Korean =
users, and wouldn't seem to make any extra difficulty for foreigners (in =
addition, that is, to pronouncing reinforced consonants!).

         A Korean phonological distinction that would be lost if this =
proposal were accepted is the one currently indicated by HANJA ("one =
character") vs. HANCHA ("Chinese character"), the latter being an =
abbreviation for HANTCHA, with the reinforced TCH.  These would both be =
written *HANJA. The foreigner would have to learn that the *J in =
*EULJIRO is like the *JJ in *JJAJANGMYEON and not like the *J in  =
*ALJIYO.  Since in this case, the rule is one of automatic reinforcement =
of T, CH, and S (but not of K or P) after L in Sino-Korean compounds, it =
is unlikely to be mastered until the language is mastered. Still, my =
feeling is that leaving a feature like this out of the romanization is =
like showing disrespect for the language.

=20

Explanation of "The Romanization of Korean" (Revised Proposal)

p. 10 =20

IV. Advantages of the Revision

* Convenience in the Informationization Age

         - By getting rid of the breve and the apostrophe, [the system] =
can be used easily on the computer.

* Resolution of Dissatisfactions of Koreans

         - The lack of distinction between voiced and voiceless =
consonants is in agreement with the pronunciation perceptions of =
Koreans.

         - Dissatisfaction with Pusan and Kwangju being pronounced =
P'usan and K'wangju will be resolved.

=20

JHTH NOTES:

         Why would a system without the apostrophe be more easily used =
on the computer? (Users of English don't seem to knock the apostrophe's =
usability on the computer.) There are even occasional complaints that it =
isn't on the keyboard, but it is, usually nowadays just to the left of =
Enter. The breve is a problem that can easily be handled by today's =
computing systems.  (MS Word 98 and 2000, I know, have fonts with =
"breved" vowels.) However, I have elsewhere suggested using the =
circumflex (even easier on a computer =97 the French use it every day, =
after all), and, in emergencies like the present e-message, a caret (^) =
following the vowel, which even typewriters can manage. Another =
possibility would be to write the O and U in italics =97 but, of course, =
in regular style within italicized words, to maintain the contrast. =
(Until you had set up some sort of keyboard shortcuts to these not only =
Roman but italicized letters, you might want to type something like OX =
and UX and then Find and Replace All in the whole document.)

         A question that would occur to both foreign and Korean users of =
the proposed system is this: if we don't distinguish voiced and =
voiceless allophones, on the grounds that they are perceived as the same =
by Koreans, then why do we distinguish the lax initial stops (*G, *D, =
and *B) and the unreleased final stops (*K, *T, and *P), which are also =
perceived as the same by Koreans? And how do these rules square with =
distinguishing *L and *R and making phonemic changes all over the place =
for the benefit of foreigners?=20

         Finally, is it really true that Koreans are bothered by a =
foreigner's light aspiration if he is guided by the spelling PUSAN but =
not by his voicing if he is guided by the spelling *BUSAN? I rather =
doubt it. No well-known language seems to have anything quite like the =
Korean lax initials, so anything the foreigner uses for them will =
probably be noticeably off one way or another until he learns the =
language exceptionally well. (After all, Koreans who have studied =
English and even used it for years are usually painfully aware that they =
have foreign accents =97 as a matter of fact, in my opinion, they are =
oversensitive on this point.) A Mandarin speaker who equates them with =
his own unaspirated series (in Pinyin written B, D, G), for example, is =
going to sound as if he is using the Korean double consonants. I think =
the problem here might be more that Koreans associate the unvoiced =
English letters with their aspirated sounds and the voiced letters with =
their unaspirated sounds, probably making the association in middle =
school, long before they've had much chance to hear the English sounds, =
and so think P'USAN when they see PUSAN.

         For all its undeniable faults, the current official system =
(like the McCune-Reischauer system on which it is based) has three clear =
and sound principles:

         1. The aim is to make it as easy as possible for foreigners =
unacquainted with the language to pronounce it well enough to be =
understood. (A secondary aim is to stand in for Han-gu^l in =
English-language publications.)=20

         2. As far as possible, consonants sounds are written as they =
are most commonly written in English (CH as in 'church' rather than as =
in 'chaos' or in 'chef,' for example). (Strong aspiration, not a =
distinctive feature of English, is shown by adding an apostrophe. Note =
that the na=EFve foreigner will not have any idea of what it stands for =
and so will ignore it, which will leave him as near to the mark as he is =
likely to get without special tuition.  Huxley in one of his novels =
refers to a lady who is the only person he knows who can pronounce the =
apostrophe in T'ang Dynasty.)

         3. As far as possible, vowel sounds are written as they are =
written in most Roman-alphabet languages other than English. (Two =
non-front and unrounded vowel sounds not suggested by any of the five =
Roman vowel letters or by established digraphs, however, are written the =
way the two roughly equivalent English "short vowels," O^ and U^, are =
sometimes written in dictionaries and the like, with the breve. Note =
that if the na=EFve foreigner is baffled by the breve, his guess at the =
value of the O or U will get him as near to the mark as he is likely to =
get without special tuition.)

         There is an old American saying, "If it ain't broke, don't fix =
it."  A corollary might be, "If your romanization system has its =
drawbacks, but the alternative proposed has almost as many, or even =
more, don't change it." Already the current system and its predecessor =
are jumbled up in the minds of most Koreans and foreigners, not to =
mention all the idiosyncratic spellings. The NAKL proposal keeps the =
useful phonetic principle of the present system (with a transliteration =
system, you might just as well ask foreigners to learn Korean) while =
going to great lengths to get rid of the breve and the apostrophe, in =
the process keeping or introducing about as many difficulties for native =
speakers as it removes. It is certainly hard to point to any gain that =
would be worth the price of revision. =20

         In particular, on this cusp of the millennia, when computing, =
one of the considerations raised, is changing so rapidly, and it looks =
relatively easy (I would guess it would only take a matter of months) to =
write a simple program able to handle conversions between Han'gu^l, a =
transliteration, and a phonetic representation, and reversibly, too, =
this can hardly be the time to spend a fortune on changing all those =
road signs, reprinting all those scholarly works and tourist guides, and =
so forth. In fact it might not cost a penny to get Hangul & Computer and =
Microsoft Korea to compete or cooperate to create such a program (if =
they don't have them already). A fraction of that road-sign budget might =
get the current system taught in the schools, the way Romaji is taught =
in Japanese schools, according to Bob Fouser. This is one education =
program that has obviously not been undertaken. Maybe we could even get =
the English-language newspapers to stop dragging their feet on the =
breves and apostrophes. They now give the impression that the words for =
"foot" and "arm" are the same, PAL =97 undoubtedly uniquely among the =
world's languages =97 and that the SHINCH'ON and SHINCH'O^N stations on =
Subway Line No. 2 have the same name, SHINCHON!  But before long =
Koreans, including school kids, will be leaving the tasks of =
romanization and deromanization to their computers, as they are already =
leaving long division to their calculators.

=20

=20

   =20

=20

=20

=20


------=_NextPart_000_0037_01BF3ACE.506C1DC0
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> 
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">SOME NOTES ON THE =
ROMANIZATION=20
PROPOSAL PUT FORWARD BY THE NATIONAL ACADEMY OF THE KOREAN LANGUAGE AT A =
HEARING=20
ON NOVEMBER 19, 1999</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <?xml:namespace =
prefix =3D o=20
ns =3D "urn:schemas-microsoft-com:office:office" /><o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">John H. T. Harvey</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal=20
style=3D"LINE-HEIGHT: normal; mso-margin-top-alt: auto; =
mso-margin-bottom-alt: auto; mso-pagination: =
widow-orphan">    =20
I attended the hearing at the kind invitation to the list, on behalf of =
NAKL, by=20
Prof. Lee Sang Oak.  I have not yet had time to study the materials =
handed=20
out at the hearing as thoroughly as I should, but as that could take =
some time,=20
and as I was the only foreigner who attended and received them, I =
thought I=20
should report my first impressions to the list and to NAKL. I'm sure the =

materials will be available from NAKL.<SPAN=20
style=3D"FONT-SIZE: 12pt"><o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">[Some Notes On These =
Notes:<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">Trying to avoid every =
conceivable=20
problem that could be caused by different operating systems, browsers, =
and word=20
processors, I have used good old CAPITALS (instead of the usual italics) =
for=20
cited romanized Korean words and letters, using a prefixed asterisk (*) =
to=20
indicate romanizations proposed by NAKL. I have written a following =
caret (^)=20
instead of the current breve ("half moon") diacritic for the obscure =
vowels in=20
INCH'O^N and KU^MGANGSAN. I have also tried to avoid any formatting that =
might=20
not get through, however desperately Word 2000 insisted on it, except =
for=20
indenting paragraphs, which I don't think can do much harm.]</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">"The Romanization of =
Korean"=20
(Revised Proposal)</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">p. 1</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">Chapter 1. The Basic =
Principles=20
of Romanization</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>1. The romanization of Korean shall follow the standard =
pronunciation of=20
Korean.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>2. No symbols shall be used other than the letters of the Roman=20
alphabet.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>3. The principle shall be "one sound =97 one letter."</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">JHTH NOTES:</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>Principle 1, to many, may seem almost too self-evident to need =
stating.=20
It should presumably be taken as referring to the adoption of phonemic =
(or=20
phonetic) representation over the alternative of transliteration (i.e.,=20
retaining the morphophonemic nature of Han-gu^l orthography) as in the =
'59 MOE=20
system and the '97 proposal of this same body. </P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>Principle 2 uses a very strict interpretation of "letters of the =
Roman=20
alphabet," one that would rule out all the accented letters of French, =
Spanish,=20
German, and most other Western European languages that use the Roman =
alphabet=20
(i.e., it rejects the breve accent mark of the present system) as well =
as ruling=20
out the apostrophe so widely used in English orthography (although =
admittedly=20
not for phonetic purposes).</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>Principle 3 seems not to be applied very strictly, at least in =
the sense=20
that this revised proposal adds two digraphs for single vowels (*EO and =
*EU, as=20
in the '59 system) to the two (AE and OE) of the current ('84) system, =
while=20
keeping the consonant digraphs (CH, NG, and double consonants). So =
perhaps this=20
principle refers instead to the fact that no distinction is made between =
the=20
voiceless allophones (K, T, P, CH) and the voiced allophones (G, D, B, =
and J) of=20
the lax stops and affricate, nor between the unpalatalized (S) and =
palatalized=20
(SH) of the lax fricative. Note, however, that a distinction is still =
made=20
between the flap (R) and the lateral (L) values of the liquid. All these =

differences are, of course, below the phonemic level =97 that is, the =
level of=20
phonetic awareness =97 for native speakers of Korean, who therefore have =

difficulty reflecting them in the romanization.<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>One can only assume that NAKL =
was afraid=20
that the R/L contrast was a much more sensitive matter than voicing and=20
palatalization, to the extent that Koreans should have to learn to make =
the=20
distinction.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><B=20
style=3D"mso-bidi-font-weight: normal"> <o:p></o:p></B></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold">pp. 1-6<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold">Chapter 2. A Glance at the=20
Romanization<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold"> <o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold">JHTH NOTES</SPAN><B=20
style=3D"mso-bidi-font-weight: normal">:<o:p></o:p></B></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><B=20
style=3D"mso-bidi-font-weight: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN><SPAN style=3D"mso-spacerun: yes"> </SPAN></B><SPAN=20
style=3D"mso-bidi-font-weight: bold">In the vowels, the only change from =
the=20
current system is the elimination of the breve accent mark by reversion =
to *EO=20
and *EU for O^ and U^, changes apparently motivated by antipathy to the =
breve=20
(or "half moon") accent mark.<SPAN style=3D"mso-spacerun: yes"> =20
</SPAN>Anticipating the revival of complaints about the failure of these =

spellings to suggest anything like the sounds intended, NAKL elsewhere =
commits=20
itself to a vigorous campaign to educate foreigners.<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>This will give them several =
hundred=20
thousand students a year, I believe, most of them fleeting=20
targets.<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>In the diphthongs, the only surprises are *WO, not *WEO, for what =
should=20
be WO^ but is actually WO in the current system and, correspondingly, =
*UI, not=20
*EUI, for U^I.<SPAN style=3D"mso-spacerun: yes">  </SPAN>Both =
simplifications=20
ignore the foreign user. It may be obvious to Korean speakers that *WO =
must be=20
*W+EO, since there is no W+O in the language, but this is hardly the =
case for=20
foreigners. In my own dialect, the former is rather like the diphthong =
in 'work'=20
while the latter would be rather like that in 'walk,' both equally =
possible. (As=20
noted, the current system has the same misguided economy.) Similarly, =
there is=20
no way for a less than fluent foreigner to know when *UI in the proposed =
system=20
stands for *EU+I and when it stands for *U+I, as in *UIDONG, unless the =
latter=20
is hyphenated, *U-IDONG. *UI does, I grant, keep *YEOUIDO down to four =
vowel=20
letters in a row! By the way, I was delighted to find HU^I romanized *HI =
in the=20
example *KWANGHIMUN. The present romanization here resorts to =
transliteration=20
rather than guiding the foreigner to the correct pronunciation. The same =
could=20
be said of the Y in cases like *TOEGYERO, I believe. At least, I don't =
hear it.=20
<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>The proposed consonant system seems to focus on eliminating the =
present=20
system's apostrophe for aspiration without using final voiced letters, =
which=20
have been criticized as sounding ugly. So Mr. PAK would be not Mr. BAG =
but Mr.=20
*BAK. This has a lot to be said for it (which in fact was said for it to =
the=20
Korean Studies list by Gary Rector a couple of years back).<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>What can't be said for it is =
that it is=20
an instantiation of the "one sound =97 one letter" principle, since it =
writes an=20
aspirated consonant the same way as the equivalent unreleased final =
stop, even=20
though the two are phonemically distinct in the language. This ignoring =
of the=20
phonetic intuition of Koreans could well introduce for them a difficulty =
very=20
similar to the one removed by not distinguishing voiced and voiceless =
plosives.=20
<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>I can't find an example in the proposal of a syllable-final stop =
followed=20
by a vowel or glide, but I presume that *BUK+AK would have to become =
*BUGAK, by=20
liaison.<SPAN style=3D"mso-spacerun: yes">  </SPAN>Otherwise you =
would have=20
BUKAK, where the first K would indicate an aspirate (although the second =
K, of=20
course, would be unreleased). This could only add to the difficulties of =
the=20
native speaker romanizer.<SPAN style=3D"mso-spacerun: yes"> =20
</SPAN>(Actually, I did find *YUN SUKYEONG, also written *YUN SUK-YEONG =
and *YUN=20
SUK YONG, but can't believe that this is intended seriously. What would =
be the=20
point of writing *SUK+YEONG differently from *SU+GYEONG, when they are=20
pronounced the same, while writing *SUK+YEONG the same as *SU+KYEONG, =
when they=20
would be pronounced differently?)<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>Reinforced J is to be written *JJ, a cosmetic change that weakens =
the=20
phonetic "meaning" of the voiced letters.<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>The reinforced consonants (the others written by doubled =
voiceless=20
letters, as at present) present another problem for the Korean user of =
the=20
system.<SPAN style=3D"mso-spacerun: yes">  </SPAN>The rules of the =
proposed=20
system would write a syllable-final stop and a following =
syllable-initial lax=20
stop as a voiceless letter followed by a voiced letter, as in the =
example=20
*GAPGOT. However, if the consonants are homorganic, that combination is=20
phonetically indistinguishable from the reinforced consonant, which is =
written=20
with a doubled voiceless letter. Compare their examples *SOKGURAM and=20
*JJILLEKKOT.<SPAN style=3D"mso-spacerun: yes">  </SPAN>Gary Rector =
nicely got=20
around this problem by writing the reinforced consonants with =
voiced/voiceless=20
pairs:<SPAN style=3D"mso-spacerun: yes">  </SPAN>KG, TD, PB, and =
TJ, a choice=20
for which he also gave an interesting phonetic justification. This =
regularity=20
might make things easier for Korean users, and wouldn't seem to make any =
extra=20
difficulty for foreigners (in addition, that is, to pronouncing =
reinforced=20
consonants!).<o:p></o:p></SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>A Korean phonological distinction that would be lost if this =
proposal=20
were accepted is the one currently indicated by HANJA ("one character") =
vs.=20
HANCHA ("Chinese character"), the latter being an abbreviation for =
HANTCHA, with=20
the reinforced TCH.<SPAN style=3D"mso-spacerun: yes">  </SPAN>These =
would=20
both be written *HANJA. The foreigner would have to learn that the *J in =

*EULJIRO is like the *JJ in *JJAJANGMYEON and not like the *J in<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>*ALJIYO.<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>Since in this case, the rule =
is one of=20
automatic reinforcement of T, CH, and S (but not of K or P) after L in=20
Sino-Korean compounds, it is unlikely to be mastered until the language =
is=20
mastered. Still, my feeling is that leaving a feature like this out of =
the=20
romanization is like showing disrespect for the language.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-bidi-font-weight: bold">Explanation of </SPAN>"The =
Romanization of=20
Korean" (Revised Proposal)</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">p. 10<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">IV. Advantages of the =

Revision</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">* Convenience in the=20
Informationization Age</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>- By getting rid of the breve and the apostrophe, [the system] =
can be=20
used easily on the computer.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">* Resolution of =
Dissatisfactions=20
of Koreans</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>- The lack of distinction between voiced and voiceless consonants =
is in=20
agreement with the pronunciation perceptions of Koreans.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>- Dissatisfaction with Pusan and Kwangju being pronounced P'usan =
and=20
K'wangju will be resolved.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal">JHTH NOTES:</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>Why would a system without the apostrophe be more easily used on =
the=20
computer? (Users of English don't seem to knock the apostrophe's =
usability on=20
the computer.) There are even occasional complaints that it isn't on the =

keyboard, but it is, usually nowadays just to the left of Enter. The =
breve is a=20
problem that can easily be handled by today's computing systems.<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>(MS Word 98 and 2000, I know, =
have fonts=20
with "breved" vowels.) However, I have elsewhere suggested using the =
circumflex=20
(even easier on a computer =97 the French use it every day, after all), =
and, in=20
emergencies like the present e-message, a caret (^) following the vowel, =
which=20
even typewriters can manage. Another possibility would be to write the O =
and U=20
in italics =97 but, of course, in regular style within italicized words, =
to=20
maintain the contrast. (Until you had set up some sort of keyboard =
shortcuts to=20
these not only Roman but italicized letters, you might want to type =
something=20
like OX and UX and then Find and Replace All in the whole document.)</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>A question that would occur to both foreign and Korean users of =
the=20
proposed system is this: if we don't distinguish voiced and voiceless=20
allophones, on the grounds that they are perceived as the same by =
Koreans, then=20
why do we distinguish the lax initial stops (*G, *D, and *B) and the =
unreleased=20
final stops (*K, *T, and *P), which are also perceived as the same by =
Koreans?=20
And how do these rules square with distinguishing *L and *R and making =
phonemic=20
changes all over the place for the benefit of foreigners? </P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>Finally, is it really true that Koreans are bothered by a =
foreigner's=20
light aspiration if he is guided by the spelling PUSAN but not by his =
voicing if=20
he is guided by the spelling *BUSAN? I rather doubt it. No well-known =
language=20
seems to have anything quite like the Korean lax initials, so anything =
the=20
foreigner uses for them will probably be noticeably off one way or =
another until=20
he learns the language exceptionally well. (After all, Koreans who have =
studied=20
English and even used it for years are usually painfully aware that they =
have=20
foreign accents =97 as a matter of fact, in my opinion, they are =
oversensitive on=20
this point.) A Mandarin speaker who equates them with his own =
unaspirated series=20
(in Pinyin written B, D, G), for example, is going to sound as if he is =
using=20
the Korean double consonants. I think the problem here might be more =
that=20
Koreans associate the unvoiced English letters with their aspirated =
sounds and=20
the voiced letters with their unaspirated sounds, probably making the=20
association in middle school, long before they've had much chance to =
hear the=20
English sounds, and so think P'USAN when they see PUSAN.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>For all its undeniable faults, the current official system (like =
the=20
McCune-Reischauer system on which it is based) has three clear and sound =

principles:</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>1. The aim is to make it as easy as possible for foreigners =
unacquainted=20
with the language to pronounce it well enough to be understood. (A =
secondary aim=20
is to stand in for Han-gu^l in English-language publications.) </P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>2. As far as possible, consonants sounds are written as they are =
most=20
commonly written in English (CH as in 'church' rather than as in 'chaos' =
or in=20
'chef,' for example). (Strong aspiration, not a distinctive feature of =
English,=20
is shown by adding an apostrophe. Note that the na=EFve foreigner will =
not have=20
any idea of what it stands for and so will ignore it, which will leave =
him as=20
near to the mark as he is likely to get without special tuition.<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>Huxley in one of his novels =
refers to a=20
lady who is the only person he knows who can pronounce the apostrophe in =
T'ang=20
Dynasty.)</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>3. As far as possible, vowel sounds are written as they are =
written in=20
most Roman-alphabet languages other than English. (Two non-front and =
unrounded=20
vowel sounds not suggested by any of the five Roman vowel letters or by=20
established digraphs, however, are written the way the two roughly =
equivalent=20
English "short vowels," O^ and U^, are sometimes written in dictionaries =
and the=20
like, with the breve. Note that if the na=EFve foreigner is baffled by =
the breve,=20
his guess at the value of the O or U will get him as near to the mark as =
he is=20
likely to get without special tuition.)</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>There is an old American saying, "If it ain't broke, don't fix =
it."<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>A corollary might be, "If your =

romanization system has its drawbacks, but the alternative proposed has =
almost=20
as many, or even more, don't change it." Already the current system and =
its=20
predecessor are jumbled up in the minds of most Koreans and foreigners, =
not to=20
mention all the idiosyncratic spellings. The NAKL proposal keeps the =
useful=20
phonetic principle of the present system (with a transliteration system, =
you=20
might just as well ask foreigners to learn Korean) while going to great =
lengths=20
to get rid of the breve and the apostrophe, in the process keeping or=20
introducing about as many difficulties for native speakers as it =
removes. It is=20
certainly hard to point to any gain that would be worth the price of=20
revision.<SPAN style=3D"mso-spacerun: yes">  </SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-tab-count: =
1">        =20
</SPAN>In particular, on this cusp of the millennia, when computing, one =
of the=20
considerations raised, is changing so rapidly, and it looks relatively =
easy (I=20
would guess it would only take a matter of months) to write a simple =
program=20
able to handle conversions between Han'gu^l, a transliteration, and a =
phonetic=20
representation, and reversibly, too, this can hardly be the time to =
spend a=20
fortune on changing all those road signs, reprinting all those scholarly =
works=20
and tourist guides, and so forth. In fact it might not cost a penny to =
get=20
Hangul & Computer and Microsoft Korea to compete or cooperate to =
create such=20
a program (if they don't have them already). A fraction of that =
road-sign budget=20
might get the current system taught in the schools, the way Romaji is =
taught in=20
Japanese schools, according to Bob Fouser. This is one education program =
that=20
has obviously not been undertaken. Maybe we could even get the =
English-language=20
newspapers to stop dragging their feet on the breves and apostrophes. =
They now=20
give the impression that the words for "foot" and "arm" are the same, =
PAL =97=20
undoubtedly uniquely among the world's languages =97 and that the =
SHINCH'ON and=20
SHINCH'O^N stations on Subway Line No. 2 have the same name, =
SHINCHON!<SPAN=20
style=3D"mso-spacerun: yes">  </SPAN>But before long Koreans, =
including=20
school kids, will be leaving the tasks of romanization and =
deromanization to=20
their computers, as they are already leaving long division to their=20
calculators.</P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-spacerun: yes">    </SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"><SPAN=20
style=3D"mso-spacerun: yes"> </SPAN></P>
<P class=3DMsoNormal style=3D"LINE-HEIGHT: normal"> <o:p></o:p></P>
<P class=3DMsoNormal=20
style=3D"LINE-HEIGHT: =
normal"> <o:p></o:p></P></FONT></DIV></BODY></HTML>

------=_NextPart_000_0037_01BF3ACE.506C1DC0--



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





More information about the Koreanstudies mailing list