There's also roumaji, which (as you presumably know but as much of the list likely does not) is the same syllabary as the kana, but represented with Latin letters - usually but not always digraphs. For example, "hiragana" or "tsunami" (the latter having been imported into English). This occasionally produces substantial changes; for example, "reiki" is a perfectly good roumaji spelling, but it leads English speakers to turn the "ei" into a "-ay-" diphthong, which is a substantial change in sound. (The word I've been spelling "roumaji" is another example; I've also seen it rendered "romaji" and "roomaji". Each is arguably inaccurate, though of course in different ways.) ...and many of the sounds the human vocal tract can make are not used (for example, the sounds English spells "th", which occurs in Japanese only when quoting words from other languages); many others are conflated (for example, the initial vowel of English "acid" and the first vowel in English "start" will generally be taken as ignorable variations on the same sound). In this Japanese is no different from any other natural language. English, for example, does not include the clicks that occur in some African languages, or, except in loanwords - where it tends to turn into a "k" sound - the guttural that Arabic, Russian, and German have (German typically spells it "ch", Russian spells it with a letter than looks rather like a Latin "X", and I don't know the Arabic alphabet enough to say anything about how it's spelled there); and English pretty much ignores the distinction between the vowels in French "rue" and "roue", leading an untrained English ear to, generally, hear them as the same vowel. I'm sure analogous remarks are true of every other natural language, though of course the details will differ widely. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B