I remember when I was in Tokyo, I never could hear に when I was asking something to a shop clerk or station staff said about the price of something. If something I heard was 百円 automatically I thought, “well, it must be 二百円 then”
Good thing they always had a calculator in hand and showed the value 
But yes, sometimes we hear people saying words differently.
Shop keepers in China always did that. They would say, “No English” and show the screen. 
Bear in mind that it’s describing a pidgin, not actual Japanese – and that it’s full of jokes! It’s poking fun at the various western merchants, chancers and other characters who turned up in Yokohama at the time when it was first opened to the West, and who conducted business in this pidgin, with I’m sure much handwaving and gesturing, and none of whom would have been learning it from a book. (My assumption is the author was writing to amuse the longer-term Western residents of Yokohama.)
(It’s also interesting linguistically because it’s practically the only information we have on the Japanese port pidgin; the trick is working out how much of it can be trusted… The eye-dialect spellings do preserve some information: the ‘ng’ in the middle of words is a pronunciation that follows the Tokyo dialect of the time, for instance.)
Not always, always exceptions.
Futatsu vs sumo, no devoicing of the futatsu versus smo wrestling.
Except they do devoice the ending -u, but it is not stuck between constanents
Others throw in your thoughts of other words as counter examples please.
Ummm… Pretty sure you gave bad examples.
sumo does not fit the pattern, も has a voiced consonant. It doesn’t fall in the category I labeled as kana that can get a tenten but doesn’t have one.
The ふ is devoiced in ふたつ in standard Japanese. You can listen to both Kyoko and Kenichi on Wanikani if you need to hear it again. The rule is relatively universal for standard Japanese. That doesn’t mean they are never pronounced (i.e, in a song or when ‘spelling out a word’), but in general, this rule is universal in standard Japanese.
A real exception might be a word like (as pointed out by Dogen) つくす where all the consonants would be naturally devoiced. Here, usually, one vowel will be pronounced (which will vary).
This rule is not so much an etymological rule, but an active pronunciation rule, so it is more or less universal. Kind of like in English, you know (assuming you speak English at a high/native level) that ed is pronounced as “t” after devoiced consonants such as in “jumped” or “kicked” and that it is pronounced as “d” as in “hummed” or “flubbed.” Even if I make up a brand verbs like fluggle and flup as in:
“I’m going to fluggle him” and later “He got fluggled.” You naturally know in English that the “ed” as a “d”.
and in
“I’m going to flup some rice” and later “I flupped some rice” you know that “ed” is pronounced as “t”.
(for what it’s worth, there is a third category in Engish when “ed” follows a dental consonant and is pronounced as “ed” as in “tilted” or “mended”)
If you find a word that’s an exception to the devoicing rule, it’s not because the word is an exception, it is because the speaker is. They are speaking nonstandard Japanese.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.