A Limitation on Names in the PRC

« previous post | next post »

Anyone who looked at the front page of the New York Times today probably noticed the article by Sharon LaFraniere entitled "Your Name's Not on Our List?  Change It, Beijing Officials Say." Featured in the article is a young woman named Ma Cheng, whose surname Ma is written with the character for "horse" and whose given name Cheng is written with a very rare character composed of three horses lined up closely in a row:  馬馬馬 (the latter character is exceedingly difficult to write in a small square exactly the same size as the space allotted to one horse [and to all other characters, even if they have as many as 64 strokes]!).  The article states that this character pronounced Cheng is not to be found among the 32,252 characters in the Chinese government's computer systems, so Ms. Ma has been told peremptorily that she must change her name. In this case, I sympathize with the Chinese government, since — as presently constituted — the Chinese script is essentially open-ended.  Anyone can dredge up an obscure, obsolete character for his / her name or invent one from scratch, and people often do just that.  Someone might even choose a character from Xu Bing's "Book from the Sky" (which contains 4,000 characters consisting of perfectly acceptable components rearranged in unprecedented ways), assign a sound and maybe a meaning to it, and claim that this is his / her name.  That way lies chaos — at least in modern society where we depend upon computerized systems for keeping all sorts of records in order. As a matter of fact, Ms. Ma really doesn't have much of a leg to stand on (no jokes about the twelve legs of the three horses in the character for her given name!).  The reason for this is that the 馬馬馬 character is listed in unabridged lexicons as simply an old form of 騁, which means exactly the same thing ("gallop; indulge in") and sounds exactly the same (CHENG3).  This 騁 is a fairly common character and is found in all modern dictionaries and computer fonts, so the government should kindly but firmly tell Ms. Ma to use 騁 instead of 馬馬馬 for her given name.  No harm done (she still even has one of the three horses to race along with).  I'm actually surprised that they haven't already made this suggestion to Ms. Ma, though perhaps they were unaware of the connection between 馬馬馬 and 騁. Now, as I indicated in an earlier post ("Zhao C:  a Man Who Lost His Name"), I'm not so sympathetic with the Chinese government when they claim that their computer systems cannot accommodate pinyin letters, since pinyin is the official romanization of the PRC and is used widely for countless purposes in the military, in medicine, in archeology, in mathematics, in physics, in education, and so forth.  I believe that the PRC authorities are being disingenuous, if not dishonest, when they say that their computers cannot handle pinyin. Incidentally, if you take the same three horses in Ms. Cheng's given name and stack one of them on top of the other two, thus 驫 (see how hard it is to make out the individual strokes in a densely composed character of this sort), then you get BIAO1 ("like a herd of galloping horses"), which is also less often read as PIAO1 ("name of a river"). Sooner or later, the Chinese government will have to adopt a similar approach to character proliferation as that taken by the Japanese government.  (Judging from their treatment of Ms. Cheng, the PRC government has decided on sooner rather than later.)  Namely, the official Japanese writing system is limited to 996 KANJI for basic education and 949 KANJI for general use, for a total of 1,945 sanctioned characters, plus — as of May 25, 1951 — 92 characters especially for use in names.  The Japanese had better be careful, however, because the list of characters approved for use in names has steadily grown during the following decades, with the current number having reached a whopping 983.  Thus, there is a strong and insistent demand on the part of some parents in Japan for more special characters that they can use to name their children.  Conversely, there is a countercurrent of more and more individuals simply writing their names in the syllabic kana, dispensing with the KANJI altogether. (A nod to Wolfgang Behr for bringing the NYT article to my attention from Switzerland at 4:20 a.m. already before I had seen this morning's newspaper here on the East Coast of the United States!)



42 Comments

  1. Bill Poser said,

    April 21, 2009 @ 4:06 pm

    The irony here is of course that the Japanese have much less of a need for extra characters with which to write their names than the Chinese do since they not only can use kana but have a much wider range of names to choose from.

  2. Mihai said,

    April 21, 2009 @ 4:08 pm

    "Namely, the official Japanese writing system is limited to 996 KANJI for basic education and 949 KANJI for general use, for a total of 1,945 sanctioned characters, plus — as of May 25, 1951 — 92 characters especially for use in names."

    That is the basic set, as they teach it in school.
    But it goes beyond that, and one is expected to continue learning more on it's own (same as the English vocabulary :-)

    There is no official restriction on characters used for names.
    And Unicode, and companies, struggle to support this:
    http://appsrv.cse.cuhk.edu.hk/~irg/irg/irg30/IRGN1435_ivs-white-paper.pdf
    http://appsrv.cse.cuhk.edu.hk/~irg/irg/irg30/IRGN1435_ivs-demo-irg30.pdf
    http://www.adobe.com/devnet/opentype/pdf/Gaiji_DeLaHunt_IUC22.pdf

    Searching for "variation selector(s)" and "Gaiji" will also uncover some interesting articles.

  3. jfruh said,

    April 21, 2009 @ 4:20 pm

    Reading articles like this make me believe that either I really don't understand how Chinese writing works, or the author doesn't, or both. You often see words referred to as "characters," as if their written form is somehow independent from the spoken meaning they're supposed to represent, which strikes me as referring to English words as "letter sequences" or something. If Ma Cheng's given name is a "character" rather than a word, how do illiterate Chinese speakers understand it?

    This blog post makes me think that I might not really be grasping something basic, though. So there can be multiple characters, with related or entirely different meanings, that represent the same sounds? Is this just a way of representing homonyms? Is the Chinese government's request equivalent to, say, an English-speaking government's request that a person not use foreign diacritical marks in their name? ("Write your name as Hass, not Haß, please.")

  4. Franz Bebop said,

    April 21, 2009 @ 4:22 pm

    Is this glyph part of the UTF-8 character set? I wonder what character set is in use in the Public Security Bureau's database.

    I also wonder where the magic number 32,252 comes from. Is this more or less than the number of Chinese characters available in UTF-8? The number 32,252 is a few hundred less than two to the fifteenth power. Is it conceivable that the database stores characters using a field 15 bytes wide?

    Other countries also have restrictions on the sorts of names people can use officially. It makes sense to examine bureaucratic rules like this by comparing them to similar rules in other countries.

  5. Mihai said,

    April 21, 2009 @ 4:33 pm

    It does not look like the character is covered by Unicode, or by GB-18030.

    See http://en.wikipedia.org/wiki/GBK#External_links
    "GBK Code Table N.B. This shows the available coding space totally populated except for 2 places, for a total of 32256 glyphs"

    It also looks like the application uses GBK, even if it is newly developed, and GB-18030 support is mandatory for all foreign companies trying to sell in China.

  6. Franz Bebop said,

    April 21, 2009 @ 5:00 pm

    @jfruh: So there can be multiple characters, with related or entirely different meanings, that represent the same sounds? Is this just a way of representing homonyms?

    The answer to your question is yes. English has many homophones, but Chinese has far more of them.

    @jfruh: Is the Chinese government's request equivalent to, say, an English-speaking government's request that a person not use foreign diacritical marks in their name? ("Write your name as Hass, not Haß, please.")

    In one respect, it is equivalent, in the sense that the request is for a novel glyph.

    But in another respect, it's not the same, since 馬馬馬 is not a foreign symbol at all. The Chinese writing system already has more than 30,000 characters and well-known customs for inventing new ones. As the article points out, human beings in China had no trouble making use of the character. They perceived it as unfamiliar but not weird or foreign. That would not be the case with Haß in an English-speaking country.

  7. Bill Poser said,

    April 21, 2009 @ 5:17 pm

    The offending character is encoded in Unicode at U+299E2. Of course, most fonts don't include it. One that does is James Kass' Code2002 font. If you can see the following character, you've got a font that includes it: 𩧢

  8. Mihai said,

    April 21, 2009 @ 5:31 pm

    U+299E2 uses the traditional form of horse (U+99AC), while the one in the article uses the simplified one (U+2A6C).

    They cannot be mixed and matched.

    For instance the Simplified-based characters in CJK Extension B are between U+299E6 and U+29A10, while the Traditional ones are from U+29867 to U+299E5.

  9. Mihai said,

    April 21, 2009 @ 5:35 pm

    Checked with a Japanese colleague: the set of characters acceptable for names are now limited to around 900 characters, so the initial article is right.
    But they still have problem with the older names (they cannot go back and ask people to change their names).
    And a newer problem: the parents can select any of the allowed characters, but are free to choose *any* pronunciation (no connections with the On or Kun).
    This seems to be quite common, and makes things really tough.

  10. Bill Poser said,

    April 21, 2009 @ 5:38 pm

    "There is no official restriction on characters used for names."

    False. And none of the linked to documents indicates otherwise. You can write your name anyway you like in many contexts, but your official, legal name must be written using approved characters, that is, either kana or an approved subset of kanji. Every Japanese citizen must be included in a family register, which is kept in the office of the ward in which the head of the household resides. Japanese law restricts the set of characters in which the names registered in the family register may be written. So long as your purpose is not fraudulent, you may write your name in some other way if you wish to, but the name in the family register is your legal name and is subject to these restrictions.

    Similary, the restrictions on the set of characters used in writing apply strictly speaking only to government documents. Non-government sources can and do use characters falling outside the official list.

  11. Bill Poser said,

    April 21, 2009 @ 5:39 pm

    U+299E2 uses the traditional form of horse (U+99AC), while the one in the article uses the simplified one (U+2A6C). [sic]

    Indeed. And which do you think Ms. Ma's grandfather found in the classical texts he searched?

    The "character in the article" really is U+299E2. The corresponding simplified character doesn't exist (yet), unless it is the one that Victor mentioned, U+9A01 骑. The obvious simplification consisting of the simplified version of U+99AC 馬, namely, U+9A6C 马, written three times in a row, has not, hitherto, existed. The version on Ms. Ma's temporary ID card is a nonce character that she and some bureaucrat negotiated. (I wouldn't be surprised if what she really would like is indeed 𩧢 but that the bureaucrat wouldn't go that far.)

  12. Bill Poser said,

    April 21, 2009 @ 5:42 pm

    And a newer problem: the parents can select any of the allowed characters, but are free to choose *any* pronunciation (no connections with the On or Kun).

    This isn't really a new problem. Characters used in personal names have traditionally often had readings not found elsewhere. The same is true of place names. This is especially true in Hokkaido where many place names are loans from Ainu. Most Japanese people I know have difficulty reading the names of places in Hokkaido unless they are well known.

  13. Ugo Lachapelle said,

    April 21, 2009 @ 8:20 pm

    To the author : Leaving aside the problem I have with your idea of that "the Chinese script [being] essentially open-ended" is a problem (I don't think that having an open-ended script is a problem, though perhaps it may be a technological challenge, but challenges are what makes our world a more interesting place to live in), you have misquoted your figures about the Japanese kanji lists. There are 1006 (not 996) kyouiku characters, and 192 (not 92) jinmeiyou characters.

    To Mihai : "And a newer problem: the parents can select any of the allowed characters, but are free to choose *any* pronunciation (no connections with the On or Kun)."

    You might be surprised to know that if you go further in studying characters, there are many special kun readings designated for use in names, even for jouyou characters. Some very popular characters can list a wide variety of readings for those names. And while they may seem chaotic and unrelated, they are, and tradition does not permit a high degree of variation as you imply. Indeed, good dictionaries (like the Kanken Kanji Jiten, available at amazon.jp with good reviews) will give you lists of these. The latest surprise that I had was 「愛(めぐみ)」 as a surname. Yet, checking in my dictionary, they listed the reading, so it's not (that) "free".

  14. jdmartinsen said,

    April 21, 2009 @ 8:41 pm

    Beijing TV profiled Ma Cheng last year (link to our translation/summary). One solution that some organizations use is to compose the character themselves, allowing it to be manipulated on their own computers (though not in the official database). Ma herself said she'd pick the name 马马马马 if forced to change.

    I find it amusing that Ma Cheng, whose name has been written using the simplified form [⿲马马马] for longer than Unicode has been around, has been using a non-existent character all this time. Ad-hoc simplification using simplified radicals from the published standards goes on all the time. The simplification standard in my 1989 dictionary lists 61 simplifications under the 马 radical (not all of them defined in the dictionary itself), while a 2000 dictionary has 68 headwords that contain the 马 radical.

    The sensible thing, I think, would be to grandfather in characters in the names of people over a certain age. This would prevent people going hog-wild over Xu Bing's imaginary creations.

  15. Bob Moore said,

    April 21, 2009 @ 9:00 pm

    Does this remind anyone else of the period during which The Artist Formerly Known as "The Artist Formerly Known as Prince" changed his name to a completely novel character, because a dispute with his recording company restricted what he could do under the name "Prince"?

  16. Chris Davis said,

    April 21, 2009 @ 9:26 pm

    @Ugo Lachapelle:

    You might be surprised to know that if you go further in studying characters, there are many special kun readings designated for use in names, even for jouyou characters. Some very popular characters can list a wide variety of readings for those names. And while they may seem chaotic and unrelated, they are, and tradition does not permit a high degree of variation as you imply.

    Tradition may not permit it, but recent trends in naming are highly nontraditional. One example I came across recently is a young girl whose name is 月. That's the character for moon. It's read "luna", as in the Latin for "moon". This is not, to put it mildly, in accordance with any established convention for how the character 月 should be read.

    A visit to any elementary school in Japan will supply you with more examples.

  17. Mr Punch said,

    April 21, 2009 @ 9:30 pm

    This story has been widely reported as "a limitation on names," but appears to be about characters. As Franz Bebop notes, many countries have restricted names to an accepted list — I recall a controversy in France about the use of traditional Breton names. In this country I assume (Prince aside) that names must be at least alphanumeric – I think of Jennifer 8. Lee of the New York Times — though I believe that the Times has inserted a period after the "8" that she didn't have when she wrote for the Boston Globe. Perhaps they think it's short for some longer number; or maybe it's a decimal point.

  18. jfruh said,

    April 21, 2009 @ 9:40 pm

    Mr. Punch — my understanding (I think Lee put a picture of her driver's license on the Internet at some point when discussing her name) is that legally her middle name is "Eight", spelled out. Maybe 8. is short for Eight?

  19. Ugo Lachapelle said,

    April 21, 2009 @ 10:00 pm

    @Chris Davis

    "Tradition may not permit it, but recent trends in naming are highly nontraditional. One example I came across recently is a young girl whose name is 月. That's the character for moon. It's read "luna", as in the Latin for "moon". This is not, to put it mildly, in accordance with any established convention for how the character 月 should be read.

    A visit to any elementary school in Japan will supply you with more examples."

    Then, if that is the case, that's something of an interesting feature.

    I don't know however its value. There has been a recent trend where I live to spell out names in odd ways. "Kelly-Anne" has been a popular name last year, there were at least 15-20 cases with that name in a popular newspaper. But none had that name spelled the same way (Kelian, Kelyann, Kelyan, Kélianne, etc.). It causes a lot of trouble for spelling : my name, a precedent in that trend ("Ugo" spelling is uncommon where I live) has always been a pain to have it written correctly. How many "I can't find you!" have I heard.

    But, in another way, having such a diversity can be an interesting feature. Japanese is already full of double and triple readings for characters, it's interesting to see that feature being expanded by knowledgeable parents.

    We'll see how it comes out to be in the future. Maybe we'll get to see and we'll think of China's move as being really backwards.

    @Mr Punch : The way I read the article, the name is actually a combination of the character and its pronounciation. The originality of the name (like in the example I gave of where I live) is in its spelling, not its sounding. China might want to limit these for technological reasons, but even if the idea might seem sound, I don't think it is (technological challenge thing and all in my earlier comment).

  20. Howard Randol Peirce said,

    April 21, 2009 @ 10:07 pm

    My name consists of three surnames — my given name is traditionally an English surname which has long been coopted as a given name, my middle name is my mother's maiden name, and my surname is a reasonably common Norman name, equivalent to English "Peter" or "Peterson," or French "Pierre." Both my middle name and my surname use relatively rare variant spellings (although the prominence of philosopher Charles S. Peirce [no relation] means that Peirce is a bit better known than variants like Piers or Pears or Pearce). It's all done with the standard roman alphabet without diacritics or novel characters (except maybe for the w in Howard, which is still used in dozens of European languages).

    No computer systems balk at my name, and in fact the unique spellings, when entered correctly in a database, make it easier to identify me and only me. There is a history behind each of my names, some of which I know and some of which I don't know, but it gives me a sense of place in time. It feels like me.

    I feel a strong affinity for my name, and my sympathies are with Mrs. Ma. It would be easy enough for the government to tell me, "Change your name to Howard Randall Pierce," but it certainly wouldn't feel the same. The modularity of the roman alphabet makes my name possible — even desirable from a data management viewpoint — but in an electronic age, I don't know what to tell 馬 馬馬馬. It's actually a very elegant and playful name written out; the conventional representation would definitely lose something.

  21. Ellen said,

    April 21, 2009 @ 10:25 pm

    Ugo is Hugo, pronounced without the H, and with a spelling variant to match?

  22. Ugo Lachapelle said,

    April 21, 2009 @ 11:27 pm

    @Ellen : I live in a French-speaking region. "Hugo" is the common form, but both "Hugo" and "Ugo" are pronounced the same : /ygo/.

  23. Philip Spaelti said,

    April 21, 2009 @ 11:45 pm

    @Ugo: Leaving aside the problem I have with your idea of that "the Chinese script [being] essentially open-ended" is a problem (I don't think that having an open-ended script is a problem, though perhaps it may be a technological challenge, but challenges are what makes our world a more interesting place to live in),

    Of course driving without airbags or seatbelts, or flying without a parachute or an oxygen mask makes life more interesting too. You should try it some time.

    A writing system is a technology for solving problems and helping people make sense of things. An open-ended writing system is a system for creating problems and making things more difficult. You could also try asking the people who have to deal with it on a daily basis — like my son — whether it's interesting.

  24. Tracy Walsh said,

    April 22, 2009 @ 12:18 am

    I was curious to see a decent version of what this character really looked like, since I don't have a font that displays it. Wikipedia has graphic at a scale large enough to see details: http://commons.wikimedia.org/wiki/File:Chinese_character_Cheng.svg

    If somebody thinks this is a bad rendering of it, please speak up and if possible find a better picture!

  25. Alex said,

    April 22, 2009 @ 1:00 am

    [(amz) gratuitous, irrelevant remark deleted here] any system which doesn't even allow for all the characters included in Unicode is pretty ridiculous — and in the article they say they're in the process of culling the 'approved' list for names from the ~30,000 they claim to support now to something closer to 8,000.

  26. Alex said,

    April 22, 2009 @ 1:01 am

    The extreme paucity of surnames in China compared to most other countries is also interesting and deserving of a post by someone much more knowledgeable than I.

  27. Lance said,

    April 22, 2009 @ 2:49 am

    There's something deeply ironic about the Chinese government saying, in essence, "this is something that would make it a lot easier if you could adopt a name just for identification purposes that's easier for the Chinese government to deal with", right around the same time Texas state rep Betty Brown ill-advisedly said the same thing.

  28. Philip Spaelti said,

    April 22, 2009 @ 3:10 am

    any system which doesn't even allow for all the characters included in Unicode is pretty ridiculous

    Really?! In what way is it ridiculous for any group — and governments are really the only kind of group with enough clout to make it likely they might succeed — to want to limit the number of symbols in daily use? I'm a big fan of Unicode. But Unicode is essentially a "garbage collection". In an effort to get everybody on board, any character, no matter how deprecated, was added to the list. That is good. But the pupose of Unicode is something else. On the other hand the idea that states, governments, or bureaucracies should bow to this as some holy standard, that they should be bound by this list, that is nuts.

  29. Brian said,

    April 22, 2009 @ 4:02 am

    When I went to the local ward office in Japan to register my son's name, they refused the Kanji we chose. It wasn't on the list because it was too old. They said we would have to use something else and copied off a page of characters from their official name Kanji book that have the same reading. When I asked if it were totally impossible, they said we could appeal the character, but it would take a few years for the process to go through.

  30. Simon Cauchi said,

    April 22, 2009 @ 4:23 am

    @Alex: The extreme paucity of surnames in China . . .

    You mean, it's worse than Wales?

  31. Nick Lamb said,

    April 22, 2009 @ 6:26 am

    The problem of storing arbitrary names in a computer database doesn't actually go away if you ban people from registering particular names as part of their legal identity. If police find that the guy running illegal gambling in their city is referred to with a symbol which resembles three concentric circles, they will want a consistent and preferably brief way to write that which can be cross-referenced with documents from other cities to find out if there is relevant intelligence from elsewhere. They won't care that Unicode doesn't have three concentric circles as a known symbol, or that the database software doesn't cater for arbitrary Unicode sequences. They especially won't care that the symbol doesn't appear on their typewriter keyboard layout.

    Deciding whether two symbols are "the same" or "similar" rather than "different" is a somewhat tricky problem, but it's one that you want to solve regardless of whether you manage to persuade your population to be identified by sequential integers or allow them to name themselves however they like.

  32. John said,

    April 22, 2009 @ 7:47 am

    This is not like the government telling you to change your name from Ugo to Hugo or Peirce to Pierce. This seems more like, as Franz Bebop said, telling you to change your name from Haß to Hass, or telling "Prince" he can't put a made-up squiggle on his driver's license. And just because they're archaic characters rather than foreign ones doesn't really make a difference, you couldn't spell an English name with OE æ, ȝ, þ or ƿ (at least not for official purposes – of course, you can spell your name in practice however you want, as in the case of Prince. I am assuming that the government isn't saying Ma Cheng is legally prohibited from referring to herself using that character, they just need a version of the name that they can spell that is legally associated with her identity.

  33. aaron said,

    April 22, 2009 @ 10:54 am

    That would not be the case with Haß in an English-speaking country.

    That name would be just fine in USA/UK, as long as you didn't mind being called Hab by folks who read your name…

    @Alex: The extreme paucity of surnames in China . . .

    I remember reading that the longer a particular culture has used surnames, the less variety there are over time. It was an interesting article; I wish I could find it with a quick google search to include here but no luck.

  34. KYL said,

    April 22, 2009 @ 11:31 am

    @Bill Poser,

    [The Japanese] not only can use kana but have a much wider range of names [than the Chinese?] to choose from.

    Bill, I'm confused by this. It's true that the Chinese have relatively few surnames, but you don't get to "choose" that anyway. But the name that you do get to choose, your given name, is essentially limitless in Chinese since the Chinese can pick any combination of characters to make up the name. So what did you mean by that part of your comment?

  35. Mihai said,

    April 22, 2009 @ 2:02 pm

    Just in case: U+299E2 is not the character in the article.
    Most likely Ms. Ma's grandfather found the real thing in classical texts and someone botched it at some point (no way to tell who).
    But reality now is that Ma is using an invented kanji for her name. So I don't have much sympathy, really.

  36. JimG said,

    April 22, 2009 @ 4:34 pm

    Further to Franz Bebop @4:22 and others, the Chinese rules aren't really more onerous and obnoxious than those of certain Hispanic Catholic countries. In some of those countries, parents were (still are?) allowed to give a newborn any name they desired, as long as it is the name of a Catholic saint. Likewise, in some societies, all females' names started with Mary. This applied to non-Catholics, including non-Christians. In Spain under Franco, Basque and Catalan names were proscribed.

  37. KYL said,

    April 22, 2009 @ 5:50 pm

    @Mihai,

    She likely decided to use the simplified form for the components because it's easier to explain to people how to write it ("write 'horse' three times" instead of "write 'horse' in traditional form three times"). As for using an invented hanzi (not an invented kanji), all such characters were invented at some point by someone, so why not her?

  38. Barbara Phillips Long said,

    April 22, 2009 @ 9:30 pm

    Is anyone tracking novelty-seeking in naming? My observations of birth announcements and honor rolls lead me to believe many parents in the U.S. seek to give their children "unique" names, and I've read some baby naming site postings that lend credence to my belief.

    I did not know Japanese parents were actively seeking novelty in naming, but it makes me wonder if there are parallels between globalization, socioeconomics and naming trends.

    I also see some parents choosing orthography that seems to have no linguistic function (apostrophes in novel first names, such as "De'andrea"). I am willing to believe that Ma Cheng or her parents just decided particular characters looked good and stuck with them; ignorance, obstinacy and whimsey seem to be equal factors in naming in some households.

  39. Nigel Greenwood said,

    April 23, 2009 @ 6:34 am

    On a slightly different topic, what is the official status of such hapax graphomena (or nonce-characters) as those used by YR Chao for "brillig", "borogroves", etc in his 1922 translation of Lewis Carroll's The Jabberwocky?

  40. David Cantor said,

    April 23, 2009 @ 10:11 am

    Regarding the disconnect of the character for a name and the pronunciation, this is also true in American English. One can pronounce one's name however one likes. It is perfectly legal to spell your name "John Jones" but pronounce it as though it were spelled "Sally Smith." Confusing to all but your closest friends, but legal.

  41. Ellen said,

    April 23, 2009 @ 10:46 am

    Looks to me like the apostophe in De'andrea does have a linguistic function. It shows there's a syllable break.

  42. Aaron Davies said,

    April 23, 2009 @ 11:34 am

    Mr Punch: one is reminded of tom lehrer's Hen3ry (the '3' is silent)

    David Cantor: one is further reminded of monty pythong's Luxury Yacht/Throatwobbler Mangrove

RSS feed for comments on this post