#1
How do Japanese dictionaries like the 大辞林 signal it?
Is it the number inside square brakets?

Quote:れい-かん【霊感
れい-かん [0] 【霊感
(1)霊的なものを感ずる不思議な気持ち。インスピレーション。
(2)神仏の不思議な感応。霊応。

Is it the [0]?

According to this link, a 0 means that the word starts with a low pitch on the first mora, and then the pitch rises from the second mora on, included the following particle.
A 1 means that the pitch start high and then it drops from the second mora on (particle included?).
A number greater than 1 means that the word starts low on the first mora, then high from the second mora until the mora indicated with the [n], and then it drops again (what about a particle following the word?).

What if there's no number at all? Does it mean that the pitch is flat?


Is it so? Thank you in advance!
Edited: 2017-11-12, 11:15 am
Reply
#2
This seems to be the explanation (in Japanese) for the Daijirin's accent markings. (I don't think the conventions are always the same across dictionaries, so it's best to check the waffle for the one you're using, which should be findable somewhere within it.) I think 'no pitch indication' simply means 'no accent info provided' -- they say they don't indicate pitch for personal names, placenames, dialect words, Buddhist and other specialist terms, etc. The waffle says that in standard Japanese every word has a pitch change between the 1st and 2nd syllables, so 'dead flat' is not an option.

I'm not sure where you're getting the entry text you quote -- in the printed Daijirin it seems to use numbers in square boxes, but the electronic version may well differ. I would suggest looking up some words that you know the pitch accent for and checking how they're shown.
Edited: 2017-11-12, 12:23 pm
Reply
#3
(2017-11-12, 11:14 am)cophnia61 Wrote: According to this link, a 0 means that the word starts with a low pitch on the first mora, and then the pitch rises from the second mora on, included the following particle.
A 1 means that the pitch start high and then it drops from the second mora on (particle included?).
A number greater than 1 means that the word starts low on the first mora, then high from the second mora until the mora indicated with the [n], and then it drops again (what about a particle following the word?).

The general rule of thumb is that particles are considered part of the word they modify for the purposes of pitch accent determination. However, some particles behave slightly differently from others. Imabi's introduction to the pitch accent mentions the differences between the various particles, although not exhaustively. It's best to show those differences by category of accent (my sources are printouts I got at a class dedicated to Japanese pronunciation at a language school in Japan; I think they're from this book, but I'm not 100% sure).

Particles attached to unaccented words (accent 0) work just as described in the Reddit thread, with several exceptions. My source claims that the only unusual 1-mora particle is ね, which becomes accented after unaccented words, although in practice it's hard for me to imagine a sentence where you would notice any difference. In most of the particles containing more than one mora (except から, ほど, きり, しか, だけ and honorifics like さん) the first mora of the particle becomes accented. Three particles can also be accented on the last mora - かね (always accented on the last mora), かな (accented on the last mora when used in a questioning way and on the first mora if used to express "being impressed"/感動, whatever that's supposed to mean) and ゆえ (can be accented on either mora interchangeably).

Particles attached to words with a non-zero accent are generally pronounced low. The major exception is の, which is pronounced high after words accented on the last mora, thus making it de facto unaccented (for example 山(や)が is accented on the second mora, but 山(やま)の is unaccented). Also, だけ can turn any word unaccented, although that is optional.

And that's just for the nouns; verbs and adjectives also have some additional particle- an conjugation-related accent rules...
Reply
Thanksgiving Sale: 30% OFF Basic, Premium & Premium PLUS Subscriptions! (Nov 13 - 22)
JapanesePod101
#4
Wow that's a very interesting and complex topic! As I'm very bad at accents in general (included my own L1 lol) I can't think how I would be able to learn this ahahah
For now I'm just trying to be able to "hear" the pitch accent, before I even try to reproduce it, and your explanation will be very useful to me in this respect, I'm sure.
So, thank you very much to both of you!
Reply
#5
Keep in mind, that just like other languages that have diverse speakers, this all varies. As an example, I believe I was once told that NHK Japanese/Tokyo accent had 雨 and 飴 (あめ) accented one way, but then in Osaka it was flipped around. So pay attention to your sources you listen to, if the Osaka dialect is heavy, then you can probably assume the pitch on some things is different from what you might see in standard dictionaries.
Reply
#6
If you are interested in pitch accent, I think Dogen's Pateron is a pretty comprehensive source. Downside, the pitch accent course is only for $10 a month subscribers. Upside, I think it's worth the money.
Reply
#7
(2017-11-12, 11:43 pm)vix86 Wrote: Keep in mind, that just like other languages that have diverse speakers, this all varies. As an example, I believe I was once told that NHK Japanese/Tokyo accent had 雨 and 飴 (あめ) accented one way, but then in Osaka it was flipped around.  So pay attention to your sources you listen to, if the Osaka dialect is heavy, then you can probably assume the pitch on some things is different from what you might see in standard dictionaries.
Yes, accent varies in different dialects -- the wikipedia article on pitch accent has a good summary with a map; there's a lot of variants beyond just 'like tokyo' or 'like kansai', including areas with no pitch accent (all words uniformly flat pitch).
(My ear for pitch accent is terrible though so this is theoretical knowledge rather than something I've actually personally noticed!)
Reply