'Denpa' should be renamed to 'Dempa'

Posted in

#1 by gambsgambs
2021-06-23 at 21:29
< report >The Japanese kana ん is pronounced as an 'm' when occurring before 'b' and 'p' sounds.

Leaving the so-called 'denpa' tag (link) written as-is will create bad habits in, and possibly irreversibly damage the pronunciation of, many aspiring Japanese learners. VNDB should take a stand in favor of correct and accurate pronunciation of Japanese, especially since there's no particular reason for the alternative
#2 by mrkew
2021-06-23 at 21:34
< report >There's this amazing page - link
#3 by mutsuki
2021-06-23 at 21:41
< report >if you read it right it sounds exactly the same
#4 by beliar
2021-06-23 at 22:13
< report >VNDB does not subscribe to this romanization. We have Senpai and Denpa, not Sempai and Dempa. And Gambs really should be the last person to lecture us about bad habits....
#5 by gambsgambs
2021-06-23 at 22:24
< report >But why? Why write it in a way that doesn't reflect its pronunciation at all, in a way that will mislead Japanese learners? You might as well write it as 'dwenpwa' if pronunciation and romanization don't have to be connected in any way; at least people will recognize that something is off then. The difference between 'denpa' and 'dempa' is subtle enough that many people may actually be confused as to how they should pronounce it
#6 by mutsuki
2021-06-23 at 22:51
< report >romanisation can be used as a formalism of the japanese sounds, not a pronounciation guide
#7 by eacil
2021-06-23 at 22:51
< report >I agree with OP, we should also rename Brocon to Blocom.
#8 by gambsgambs
2021-06-23 at 23:08
< report >#7 it's only pronounced as an 'm' before 'b' and 'p', and you can't recognize that because VNDB 'subscribes' to a fraudulent romanization system. This is exactly the problem I'm bringing up
#9 by myopius
2021-06-24 at 01:21
< report >
The Japanese kana ん is pronounced as an 'm' when occurring before 'b' and 'p' sounds.
Rather than saying that ん is pronounced as "m", it's more accurate to say that Japanese ん lacks the hard stop of English "n" which prevents English "n" from seamlessly transitioning into "b" or "p" (which begin the same way "m" ends: lips closed) in such a way that you hear an "m" sound.

Even if VNDB romanized ん as "m" in these cases, it's not it would result in a romanization system that causes English speakers to pronounce Japanese like a native. (If you think that the Japanese "r" sounds exactly like the English "r", go watch more anime.)

Rather than make a half-hearted attempt to teach English speakers to pronounce Japanese better, VNDB utilizes a romanization system that has symmetric translation between kana and English letters. Simple and straightforward.

tl;dr Sorry, VNDB never promised to teach its users to speak Japanese properly.
#10 by gambsgambs
2021-06-24 at 02:17
< report >
VNDB utilizes a romanization system that has symmetric translation between kana and English letters. Simple and straightforward.

This isn't true because 'つづく' is rendered as 'tsuzuku' on this website -- it's indistinguishable from 'つずく'
#11 by naiohoras
2021-06-24 at 02:51
< report >even the Japanese doesn't stick to its actual kana pronounciation. damn the Japanese, right? making us foreigners (and themselves) confused.Last modified on 2021-06-24 at 02:52
#12 by kametec
2021-06-24 at 04:33
< report >I'd like to raise two points regarding OP.

First is a link to FAQ, which says:
we have decided to use a mix of Hepburn and Wapuro for all fields. This is mostly consistent with AniDB's romanization.

Modified Hepburn seems to be used. So instead of discussing whether a single word should be changed, you should include this change in a proposal to change whole romanization system. It is of little use to make changes on per-word basis.

Second, learners of Japanese can generally read kana and use some other resources than vndb tag romanization for their pronounciation needs.

Oh, and I cannot help but add a third in reaction to #10: つづく and its romanization is an anomaly. It just quietly exists and doesn't cause any major confusions, because learners learn about it one way or another. I'm also yet to see a context where つずく would be used.
#13 by adamstan
2021-06-24 at 05:00
< report >
I'm also yet to see a context where つずく would be used.

According to jisho, there's no such word - it's only used as a reading for some names ;)
So, while it's just one dictionary, it suggests that even if such word exists, it's rather obscure.Last modified on 2021-06-24 at 05:01
#14 by vninfohata
2021-06-24 at 08:31
< report >Ignore this troll for your sanity is more important.
#15 by gambsgambs
2021-06-24 at 08:33
< report >I’m just saying that づ and ず both seem to be rendered as “zu” here so the bizarre romanization choices don’t even form a bijection between kana and their Roman forms as someone had suggested is the case. つずく isn’t a word
#16 by arkt
2021-06-24 at 09:29
< report >I have N1 and hate when people use m instead of n.
#17 by adamstan
2021-06-24 at 09:31
< report >Yeah, I found it really weird when Mangagamer wrote "Sempai" in their older translations.
#18 by vninfohata
2021-06-25 at 00:47
< report >Petition to rename Gambsgambs to ganbsganbs. We did it already on our server link
#19 by historyeraser
2021-06-25 at 02:41
< report >What server?
#20 by luckydaikon
2021-06-26 at 15:42
< report >
written as-is will create bad habits in, and possibly irreversibly damage the pronunciation of, many aspiring Japanese learners.

I can't say I understand the logic behind this. You are going to have to learn that pronunciation rule eventually anyway. Let's say you are reading Japanese and hit the word こんぶ. Let's pretend you don't know this word. You can instantly tell it is pronounced kombu, but not because you saw it in romaji once. You know it because you have memorized the pronunciation rule.

If anything is going to damage your Japanese pronunciation (not irreversibly, I think you can correct most mistakes with practice), it's relying on romaji any longer than you absolutely have to.

bijection between kana and their Roman forms
Sometimes you just aren't going to have this. Every sound in every language isn't going to match up in a 1=1 way to English, that's why the International Phonetic Alphabet was invented. Romanazation is an approximation.Last modified on 2021-06-26 at 15:50
#21 by ninius
2021-06-26 at 15:51
< report >In Finnish there's a word "niinpä" that's very often pronounced "niimpä" and some clowns even write it so. This doesn't change the fact that the word is still officially "niinpä". Just saying.

PS. "Sempai" is fucking cringe.Last modified on 2021-06-26 at 15:54
#22 by kiru
2021-06-29 at 09:33
< report >@9 already said how it is. But romanization isn't about how you speak it, otherwise the whole r/l, ryu and so on things would be quite different as well. Senpai makes more sense to use, as it's technically an n anyway. Even if it ends up becoming an m.

Wanting to have something being spoken the same in English would cause changes like Jakuri->Jacqli that Jakuri had. That has nothing to do with romanization anymore.
#23 by kingsley
2021-06-29 at 10:56
< report >this post is obviously pointless, as some language romanization isn't always be meant to be the way you speak, as how people here have explained. Also I can bet this OP isn't even familiar with Japanese language as a whole, like the word "Senpai" and "Denpa". This kind of words always remind me of how you say things in Arabic, like how certain consonant sounds like when being brought together into one word. Surely everyone can ignore the way it sounds like if you're a non-native, but complaining (or whatever this post is) isn't gonna change anything.
#24 by travtownley
2021-06-30 at 12:00
< report >Well, this might be another debate between Prescriptivism (language must follow the rules) and Descriptivism (language should follow the official one or what's widely used).

I'm sure that VNDB is a prescriptivist because of the guidelines here: link. Personally, I value prescriptivism since using direct romanization (i.e. 遠坂、とおさか --> toosaka) is better than any romanization which "alters things" mid-romanization (e.g. 遠坂、とおさか --> tohsaka) so that I can return the romanized letter back to the Japanese letter (toosaka --> とおさか) without resulting in an invalid result when it is re-romanized (tohsaka --> とふさか --> tofsaka, the tofsaka is the invalid one, considering ふ can be both fu and hu)
#25 by tester
2021-06-30 at 12:13
< report >
so that I can return the romanized letter back to the Japanese letter without resulting in an invalid result when it is re-romanized

Well, you can do that 100% correctly (I'm not talking just about cases like "とお" or "こお") only via... yes, good old Nihon-siki.Last modified on 2021-06-30 at 12:39