Sunday, December 05, 2010
a member of my family circulated the above video by email this weekend. it's just a joke, but it happened to trigger my pet issue about how the voiceless velar fricative should be transcribed in english. yeah, i know, everyone has strong feelings about this issue. but actually, i think the problem is fairly straightforward (although it will take me a little while to set it up).
transliteration is essentially trying to spell something in one language using the alphabet used for a different language. the purpose of transliteration is to make people who don't speak the transliterated language be able to come close to pronouncing the foreign word. sometimes transliteration is easy, because sometimes each sound in the original language has an exact equivalent in the target alphabet. so "גט" is transcribed as "get". and as it happens, the hebrew pronunciation of the original word is pretty much like the english word "get", which makes it an easy word to transcribe.
the harder issue is when you're trying to transliterate a sound that doesn't exist in the language of the target alphabet. because the transcription business is fairly established, there are a lot of conventions that you can fall back on to represent sounds in an alphabet even though the sound isn't ordinarily represented in the alphabet. often that is done through a combination of letters. "gh" for example, is often used to represent the arabic "غ", which is a voiced velar fricative, a sound that does not exist in english. so "baghdad" is not really "bag" + "dad" (as most english speakers say), it is "ba(gh)dad", where the (gh) itself is a single sound that is kind of like the rolled "r" in french. the reason "gh" is chosen is because: (1) that combination of letters hasn't already been assigned a sound, and (2) because even if it is pronounced like a "g" and "h" by an ignorant english speaking, it's close enough for a native of the original language to figure it out.
which brings me back to the voiceless velar fricative, another sound that isn't used in most standard dialects of english. but it's used in a whole bunch of other languages. it is the "خ" in arabic, the "Х" in russian, and both the "ח" and the "כ" in hebrew. the "ח" is the first letter in the hebrew word "hanukkah" (חנוכה).
for some reason the "ח" often gets transcribed in english as a "ch". that's why "hanukkah" is sometimes spelled "chanukah". it drives me crazy because using a "ch" breaks one of the rules of transcription: avoid using letter combinations that already stand for a different sound in the transcribed-to language. "ch" already stands for a sound in english (the voiceless postalveolar affricate), a sound that is completely different from the one that the "ח" makes in hebrew. by spelling the holiday "chanukah", it looks like the first sound in the word is like the first sound in the word "church". that is misleading. and also it makes little sense because when the same sound appears in other languages, that sound is usually transcribed as "kh". "kh" is much better. it's a combination that does not exist in english (except in foreign transcribed words trying to represent that sound, like at the end of "kazakh"). it just seems like a better candidate, one which is successfully used in other languages and doesn't have the potential of misleading native english speakers into thinking it is a "ch" as in "church" sound.
so in the debate whether to spell the english word for the jewish festival of lights "hanukkah" or "chanukah", i think they're both wrong and it should be "khanukah".