ROBBIE LYMAN

One of the things that fascinates me about language is the degree to which the written representation of language does or does not capture the pronunciation. On the one hand you have French, which I don't speak with any skill at all, but do sometimes marvel at comparing text to what I understand to be the correct pronunciation of it. American English of all stripes does this too, of course: easy to imagine somebody tripping over the correct-to-locals pronunciation of placenames like Delhi (NY), Worcester (MA), Wayzata (MN), or Houston St (in Manhattan). On the other you have things like the Hepburn system for romanization of Japanese.

Japanese, officially, has about 50 "main" syllables (each one represented in hiragana or katakana by a single symbol) and perhaps another additional 20 "contractions", which get two ("sha" しゃ is written like "shiya" しや, except the "ya" is smaller) along with a couple variants of the main syllables ("ha" は vs "ba" ば and "pa" ぱ, for instance). Each syllable with the exception of ん (written as "n" by Hepburn, but maybe "ng" would have been closer to the sound) comprises what native English speakers would call one vowel, usually preceded by one consonant.

Hepburn's romanization system, like others, typically represents each syllable (the Japanese term is mora) this way: one can (and people do) group the kana syllabaries as a five-by-ten table (plus one extra, the ん) with the vowels (typically in the order "a i u e o") in one dimension and the consonants in the other. So one has かきくけこ "ka ki ku ke ko". One feature (possibly explanatory of its enduring popularity) of Hepburn's system is that it plays to the strengths of native speakers of English by writing, for example ち as "chi" rather than what a rote rule-following would have produced as "ti—to English speakers "chi" is much closer to the actual sound you'd hear.

Anyway, one of the fun aspects of Japanese orthography and phonology is that it has very obviously semantically meaningful glottal stops, the sound you often make in the middle of the word "uh-oh". In Hepburn, these are consistently represented as double consonants. So in mecha anime like Code Geass or Neon Genesis Evangelion, when they announce the launch (発進) of a mecha, the word they use (はっしん "hasshin") is correctly pronounced by starting to say "hosh" (long "a" as in "father" rather than the word "hash" in "hash browns"), grinding briefly to a halt, and then finishing with "shing". In Japanese orthography, this glottal stop is represented with a small version of つ "tsu".

So when I listen to King Crimson's Discipline album, I sometimes cringe a little as Adrian Belew sings the title phrase of "Matte Kudasai" (待って下さい, literally "please wait", but the Japanese phrase might deserve the dramatics Adrian lends it—for me the English somehow it evokes call centers or fast food lines but the Japanese does not): although he doesn't make the mistake of saying "mate" as in "pal" or "friend", he sounds a little funny to me for not putting the little pause between "ma" and "te". Granted, then I cringe at myself for cringing—it's a great song, and it's the kind of nitpick only a pedant like yours truly would make.

In this particular case, Adrian gets off fine—there isn't any way to sensibly interpret his saying まて when he means まって as anything other than a mispronunciation. Sometimes though missing the stop totally changes the meaning. Perhaps the most common pair of words related this way are conjugations of "to do" and "to think": して vs 知って ("shite" and "shitte". Yes, yes, I know, it is funny the first time. But really you pronounce the first one more like "shteh" and the second similarly but with the stop).

Anyway this morning I discovered that although English doesn't formalize these semantically meaningful glottal stops into its writing system in the same way, it does have them! Here's how I realized: imagine that you, like me, have said the phrase "self-own" to a friend. When you discover that they heard "cellphone", try to explain the phrase you actually said.