Japanese: Some questions about pitch accent

[RubiksKid]

I recently learned about the importance of pitch accent in Japanese (yeah, a little late), and I have some questions on the subject. I tried posting these on HiNative, but I think it's a subject that natives don't really consider, so I haven't gotten any replies.

1) Is pitch accent consistent between transitive/intransitive pairs? For example, I know that 生える and 生やす both have the same pitch pattern: low, high, low. Are patterns between transitive/intransitive pairs consistent like this in general?

2) Does pitch accent change based on context? For instance, 年 is pronounced as "とし", low, high. Is 年 pronounced the same way in 年上 and 年下? As far as I can tell, it is. What about other cases? Does this hold true in general?

3) Are there significant regional differences in pitch accent? (low, high vs high, low, etc.)

(digressing a bit) I'm currently in the process of cross referencing my vocabulary against a pitch accent dictionary and including notes (color notation) in my Anki cards.

I appreciate the help.

-Nick

[vonPeterhof]

I'm not really sure about the first question (I think that's the case, but I don't really have sources or data to tell for sure), but I can try answering the other two.

2) Does pitch accent change based on context? For instance, 年 is pronounced as "とし", low, high. Is 年 pronounced the same way in 年上 and 年下? As far as I can tell, it is. What about other cases? Does this hold true in general?

Unfortunately, it does. According to my accent dictionary 年上 and 年下 are actually accent zero, and so is お年 (as in 良いお年を). What's more, even some particles may affect it in some cases. You probably already know that the key difference between words accented on the last mora and accent zero words is the fact that the following particles don't have a high pitch, but that doesn't apply to some particles: it's としは and としを, but としの and としだけ.

3) Are there significant regional differences in pitch accent? (low, high vs high, low, etc.)

Oooh yes there are As you can see on the map, most of the country has Tokyo-style pitch, which means that A) the first and the second mora must have a difference in pitch, B) unaccented words start low and continue at an even height, and C) accented words feature a single downstep. That doesn't necessarily mean that all words in all those dialects will have the same pitch patterns though. The most common alternative style of pitch, and the only non-standard one to have a significant presence in national media, is the Keihan- or Kansai-style, where the patterns are a lot more variegated: unaccented words can have uniformly low or uniformly high pitch, a word can contain an upstep instead of a downstep, plus the actual positions of the accents tend to be noticeably different from standard Japanese. Other dialects may have zero accent in all words, accent in the same position in all accented words, or... whatever the heck that goes on in the Fukui dialect (I like to describe it as "Kansai grammar with a Seoul intonation").

While schools throughout the country teach standard Japanese grammar, as far as I'm aware the only people who receive dedicated instruction in standard (i.e. Tokyo) pronunciation are those who aspire to be teachers, actors and announcers. However, it does seem like certain dialects are getting displaced by standard Japanese with the help of the media, and in general native speakers of non-standard dialects will make their best effort to speak standard Japanese to people from outside their region (although people from Kansai and Osaka in particular are stereotyped as not even bothering to do that).

[RubiksKid]

Thank you so much for your post! I appreciate you taking the time to go into so much detail.

As you can see on the map, most of the country has Tokyo-style pitch, which means that A) the first and the second mora must have a difference in pitch, B) unaccented words start low and continue at an even height, and C) accented words feature a single downstep.

Yeah, those are the guidelines I read when studying pitch accent. I didn't realize that it was considered a Tokyo-style pitch though. I just thought it was a universal quality.

According to my accent dictionary 年上 and 年下 are actually accent zero, and so is お年 (as in 良いお年を). What's more, even some particles may affect it in some cases. You probably already know that the key difference between accent zero words and words accented on the last mora is the fact that the following particles don't have a high pitch, but that doesn't apply to some particles: it's としは and としを, but としの and としだけ.

Yikes, that's pretty intimidating. And I'm not familiar with the term "accent zero". Are you referring to the "flat pattern", as in 私? According to my studies, the flat pattern does include a difference in pitch between the first two morae, but the difference is less pronounced compared to a downstep.

Also, what accent dictionary do you use? Is it online by chance? I've been using this.

[devilyoudont]

1) Is pitch accent consistent between transitive/intransitive pairs? For example, I know that 生える and 生やす both have the same pitch pattern: low, high, low. Are patterns between transitive/intransitive pairs consistent like this in general?

Overwhelmingly yes, but with some exceptions. The common pair that has a different pitch accent is 入る and 入れる.

[RubiksKid]

However, it does seem like certain dialects are getting displaced by standard Japanese with the help of the media, and in general native speakers of non-standard dialects will make their best effort to speak standard Japanese to people from outside their region (although people from Kansai and Osaka in particular are stereotyped as not even bothering to do that).

By "standard Japanese" are you referring to the Tokyo-style pitch?

1) Is pitch accent consistent between transitive/intransitive pairs? For example, I know that 生える and 生やす both have the same pitch pattern: low, high, low. Are patterns between transitive/intransitive pairs consistent like this in general?

Overwhelmingly yes, but with some exceptions. The common pair that has a different pitch accent is 入る and 入れる.

Oh, okay. I appreciate the response. It's good to know that that's the case in general, even if there are exceptions.

[vonPeterhof]

As you can see on the map, most of the country has Tokyo-style pitch, which means that A) the first and the second mora must have a difference in pitch, B) unaccented words start low and continue at an even height, and C) accented words feature a single downstep.

Yeah, those are the guidelines I read when studying pitch accent. I didn't realize that it was considered a Tokyo-style pitch though. I just thought it was a universal quality.

Well, the vast majority of resources for learning pitch accent (as opposed to just describing it from a linguistic standpoint) teaches the standard Tokyo-based pitch accent by default. I'm not sure if I made it clear enough in my previous post, but just because an area is classified as having "Tokyo-style pitch" doesn't mean that the placement of the accent in specific words will be identical to that of the Tokyo area. I recall seeing an interview with the voice actress Shiori Mikami, who is a native speaker of the notoriously impenetrable Tsugaru dialect, and she mentioned that she has to use an accent dictionary to get the standard pronunciation right, even though her native area is classified as having an "outer Tokyo-style accent".

And I'm not familiar with the term "accent zero". Are you referring to the "flat pattern", as in 私? According to my studies, the flat pattern does include a difference in pitch between the first two morae, but the difference is less pronounced compared to a downstep.

Yeah, I meant the flat pattern. Many dictionaries mark accent using the number of the mora before the downstep, and the lack of a downstep is marked with a zero, which is why it's sometimes called "accent zero".

Also, what accent dictionary do you use? Is it online by chance? I've been using this.

I use the NHK dictionary app, which is unfortunately no longer supported. But the OJAD is a very useful resource, especially for things like accent shifts during verb conjugation.

However, it does seem like certain dialects are getting displaced by standard Japanese with the help of the media, and in general native speakers of non-standard dialects will make their best effort to speak standard Japanese to people from outside their region (although people from Kansai and Osaka in particular are stereotyped as not even bothering to do that).

By "standard Japanese" are you referring to the Tokyo-style pitch?

In that paragraph I was referring to the whole package of grammar, vocabulary and pronunciation.

[RubiksKid]

I use the NHK dictionary app, which is unfortunately no longer supported. But the OJAD is a very useful resource, especially for things like accent shifts during verb conjugation.

Oh, okay. I actually found a link to the NHK app before posting this, so that's what it seemed like (no longer being supported). And whenever OJAD doesn't have a particular entry, I'm just using Forvo and then adding the notation myself. I've tested agreement between both sources quite a few times, and it's very consistent.

I'm currently in the process of scrupulously going through about 2,500 words, providing pitch accent notation for each. The process is about as fun as it sounds. I'm seeing a lot of patterns though, which is useful. For instance, adjectives that end in ~しい usually have a downstep on the [し]い, loanwords usually have a downstep on the first mora, accent zero is the most common for native words, try-syllabic adjectives often have a downstep on the second mora, etc.

In that paragraph I was referring to the whole package of grammar, vocabulary and pronunciation.

Right, I understand. I have a question though. I know pitch accent varies by region, but insofar as a "standard" pitch accent exists, would it be the Tokyo style?

[vonPeterhof]

I actually found a link to the NHK app before posting this, so that's what it seemed like (no longer being supported). And whenever OJAD doesn't have a particular entry, I'm just using Forvo and then adding the notation myself. I've tested agreement between both sources quite a few times, and it's very consistent.

The Daijirin is also a good dictionary with pitch accent data, indicated as numbers in square brackets.

I'm currently in the process of scrupulously going through about 2,500 words, providing pitch accent notation for each. The process is about as fun as it sounds. I'm seeing a lot of patterns though, which is useful. For instance, adjectives that end in ~しい usually have a downstep on the [し]い, loanwords usually have a downstep on the first mora, accent zero is the most common for native words, try-syllabic adjectives often have a downstep on the second mora, etc.

The Imabi introduction to pitch accent has quite a few more of those (I was going to include it and Daijirin in my previous post, but I was in a hurry when I wrote it and then forgot about it). There's also this website with some useful information on pitch in proper nouns.

I know pitch accent varies by region, but insofar as a "standard" pitch accent exists, would it be the Tokyo style?

Not just Tokyo style, but specifically the pitch accent of the modern Tokyo dialect (historically the Yamanote dialect, but nowadays more or less coterminous with the speech of the non-rural parts of southern Kanto).

[RubiksKid]

The Daijirin is also a good dictionary with pitch accent data, indicated as numbers in square brackets.

Awesome. I'll add it to my list of resources. I've currently gone through about 700 words or so (~2 hrs/day).

Not just Tokyo style, but specifically the pitch accent of the modern Tokyo dialect (historically the Yamanote dialect, but nowadays more or less coterminous with the speech of the non-rural parts of southern Kanto).

Okay, I see. Thank you for the clarification.