I’m very sorry to make a post about pitch accent, haha. I’ve seen online how this can be a heated topic. To try to preempt this – I’m not looking for discussion of how useful or important it is or analysis of the people who teach it or methods of learning it. I’m very new to learning Japanese, and I’m not going to bog myself down in treating it as a crucial thing to nail right now. That said, I’m interested, and figure the earlier the better for just recognizing it’s a thing that exists, generally being aware of the patterns, and trying to hear that they are happening when natives speak. The more dedicated work can come some months down the line if I decide I want to pursue it.
All that said… I don’t think I can even hear it. For context, I’m almost thirty (I know pitch gets trickier as you age) and have had next to no musical training/education. I’ve always considered myself “tone-deaf” in the colloquial sense. I can certainly hear a difference between two piano keys or something, but I have a hard time learning to, say, map a sound onto a note or deal with more subtle differences. And boy do pitch accent differences seem subtle.
It seems to me that most methods for learning this more or less start by presenting what the 4 patterns are and giving examples of words using them (on the word level – I know sentence pitch exists but I haven’t looked further into that yet). My problem, then, is that even in these examples… they just sound flat to me. It’s all the same in my ear. If I’m presented with an on screen display, say like what I’ve seen Dogen do, I have times when I THINK I’m hearing it? But it’s subtle even then and it’s hard not to think my brain is just filling in the information for me because it’s being told the “answer.” Take away those displays and those same exact words go back to being flat. I don’t know… would it work to just drill through words with displays and try to train my ear that way? This loops back to me not wanting to spend TOO much time on pitch accent right now because I have bigger fish to fry, but it feels like where a lot of people might be starting at zero, I’m somewhere in a negative number.
I get that it’s unfamiliar to deal with pitch in this context, but I feel like my ears haven’t learned to discriminate well enough to hear it even when all of my attention is purely attuned to pitch. Is this just a normal starting experience? This isn’t me giving up or saying it’s definitely impossible for me; I just need guidance. Are there tools out there for someone who might need to start at practicing distinguishing close-ish pitches AT ALL before applying it to words like this?
So… I can easily hear that they’re not being said the same exact way. I think that’s good news. It’s almost like it doesn’t sound like a pitch thing to me, like it’s a more English emphasis. If I was just presented this without outside info, I would think the higher sounds were almost maybe longer/louder? I guess my brain is just processing it weirdly, like the way it normally thinks about language. Then again, for the actual pitch change / “emphasis,” on rain it’s very noticeable, but I’d be more hard pressed to say if candy was going up or just flat.
Just to be clear, nothing in Standard Japanese is literally flat (all low or all high pitch) and there is a category referred to as “flat” (first low and all remaining high, even including particles).
When you say if you can’t tell if candy is flat, I’m guessing you meant that you can’t tell that it’s not two low or two high ones, right? That is, you can’t hear any pitch change.
With はし there is another one, “edge”, which is flat, but since it’s two mora you need a particle to be able to tell it apart from bridge. But I’m guessing you weren’t referring to this kind of difference.
Just earlier today, I was wondering if this about myself impacts my (lack of) recognition of pitch. I can tell that あめ and あめ sound different, for example, but that’s about it. I’ll put this thread on notification so I can follow along =D
Well, what I meant to say is – of course in real usage I’m not going to ignore the patterns, but in a sort of outside “what does my ear hear” way without consciously working to map it onto something, the last one is in a zone where the rising at the end is subtle enough that I’d be stuck between guessing if it exists or not. I guess like you said, it would be easy for me to think it’s just two low. Again, purely hypothetically.
Of course, I could be totally misunderstanding stuff too. Anything relating to pitch or general… audio/music stuff is about as far out of my comfort zone as it gets, heh.
That is probably because you’re used to stress accent in English, where accented syllables are louder and longer, in addition to being higher pitch. This might actually be a good sign, since it seems to indicate that maybe your brain is interpreting an accent being there. In Japanese though, accent is expressed almost entirely through pitch, without lengthening or increased volume. Obviously people will sometimes speak louder or draw things out for emphasis, but it’s not part of the accent system in the language the same way it is in English.
Try saying “AH-meh” and “ah-MEH” out loud, stressing them how you would in English. Then listen to the clips and try to compare if they sound the same. Maybe record yourself and compare them. You should be able to hear that in the Japanese clips, the syllables are said at roughly the same volume and length as each other, especially when put right next to hearing them pronounced with English stress.
To be clear, I’ve had quite a bit of musical education and I speak a tonal language (Mandarin), so perhaps I’m completely incapable of understanding your experience, but I personally wouldn’t even start with ‘correct pitch’ in Japanese words. Do you have any music you like or any songs you’d gladly listen to on repeat, and can you hear any pitch variation in those? (E.g. If I asked you to listen to ‘Can You Feel the Love Tonight?’, would you be able to hear how the word ‘can’ at the beginning of the chorus is higher than the word sung just before it?) Try seeing what you make of those now, and ideally, pick one that you’d like to sing, if only in private. Even people with a good sense of pitch can make mistakes or get confused by what they hear, but if you learn to generate pitch variations yourself, I believe you’ll have an easier time knowing when something has changed, and in which direction, if only by imitating the sound yourself and feeling how your body changes (notably how your throat feels) as you change the pitch. Typically, especially without training, the higher we go when singing, the more tension we feel in our throats. If you can learn to identify those feelings, and then imitate how Japanese words are pronounced while applying your new knowledge, and ideally while looking at a pitch accent guide (e.g. ‘this mora is high, so my throat should feel a little tighter’), I think you’ll have an easier time.
By the way, it’s quite natural for us to use a little more air and force when we generate higher pitches, so it’s quite normal for the higher pitched morae to sound slightly louder than the others, even though it’s not necessary for the pitch accent to be correct.
Yeah that makes sense. I’ll mess around with what you suggested and see if I can force my brain to figure out the difference, thank you! It is still definitely a lot harder in words where I don’t have this point of comparison, but hey, baby steps.
Yeah! I guess I had a bit of an advantage from starting younger, but I know I also had some issue figuring this out as well. For me at first, I could tell when things sounded wrong, but I often couldn’t tell why they were wrong or how to fix the problem so that I could pronounce things better. It’s definitely a process. Knowing is half the battle, as the saying goes. Just knowing what to listen for can help you focus on picking out the differences.
Thanks! That is the sort of idea I was wondering about – since it’s hard for me to figure it out in the language, maybe spending some time with something simpler outside the language that would train the general pitch distinguishing skill. I assume that’s fairly transferable anyway. I will try to work on it more myself. I think definitely in the example song you mentioned, that one is clear enough. It’s not like I can’t hear pitch differences at all, it just feels to me like it’s a matter of how close together they can be before it gets really hard. And I think, to my untrained ear, the difference is larger there than it is on average in Japanese words? But you’d know a lot more about that than me, maybe I’m wrong and singing is just easier to hear it in than speech (for English speakers) or something.
I do think, or hope, it’s already been a little beneficial to compare words with different pitch accent and at least see that I both CAN hear it in those cases, even if only partially correctly, but also that there’s the chance I can be noticing it without really realizing I’m noticing it, if my brain just sorta conceptualizes of pitch incorrectly from its lack of training and English-centric defaults.
I recommend downloading the WaniKani pitch info script, which puts pitch information on the WK lesson and review pages next to the audio buttons. Even if you don’t understand exactly what’s going on with the pitch, if you look at the pitch pattern diagram while listening to the audio for each new vocab word you learn, it might help you slowly start to gain an understanding of how it affects the sound.
It probably is larger: I think the difference in most Japanese words is about a musical third at most, whereas that’s a fourth, so the interval is a bit bigger. And yes, it’s probably easier to catch in singing, but ultimately, the way we generate pitch changes is about the same in either case.
The reason I suggested playing around with songs first is that a song instinctively feels wrong when you hear a down-step instead of an up-step (because you’re used to hearing it one way and not the other), and I think the biggest challenge for people learning to handle pitch is always direction rather than just hearing a difference: most people can hear one, but knowing where the pitch is going isn’t always as obvious.
While this obviously isn’t the correct way to pronounce a Japanese word, I think that one thing you can try in order to make pitch changes more obvious in Japanese words is to try lengthening each mora you pronounce: drag it out while trying not to change the pitch (which you can generally do just by not changing the position of any of your mouth and throat muscles, as though you’re trying to freeze the sound), then move on to the next mora and do the same thing. You should be able to feel the amount of tension in your throat changing as you do it. For example, when I say A-me (雨, rain), I feel a little relief as I come down to ‘me’, because there’s less tension in my throat. On the other hand, when I say a-ME (飴, a sweet), I feel more tension on ‘ME’, and my throat tightens up slightly at the end of the word. You might also feel as though the air coming out from your lungs is pushing up harder when you’re pronouncing a higher pitch, and that you’re making a little more effort. That’s normal.
(An account is needed, but you can register with anything, even a fake email is ok)
I’m also really bad at hearing pitch but I typically score more than 50%, so it looks like my brain is picking up something, even when I don’t consciously hear it.
…doing substantially worse than random chance is partially what inspired me to make this topic, haha.
Actually, on that subject, maybe I’m being silly, but would someone mind explaining exactly how I’m meant to interpret the \ in the answers? I figured that was the point where the pitch changed, but at times I think it was at the end of a word IIRC. Does that mean it’s going to change on the particle? And, as someone pointed out here, don’t all words have SOME pitch change? Do the ones without a line mean the “flat but not literally flat” accent?
The pitch always falls after the accented mora. When that comes at the end of a word, that means the pitch drop comes with the particle, as you guessed!
If the accent is on the first mora, then the pitch starts high and immediately falls after the first mora. In all other cases, the pitch starts low and rises as you move onto the second mora and then falls again where the accent is. Unaccented words do not have an accent (pitch drop), so the pitch starts low and remains high through the end of the word and any attached particle.
There are some exceptions in terms of certain particles or conjugations changing a word’s normal pitch accent, but those are the basic mechanics.
So an unaccented word like 東京 will be とうきょう (L H H H H)
If the accent is at the start of a word, like き^る (to cut) it will be (H L)
If the accent is in the middle of the word, like 壊す（こわ^す） it will be (L H L)
If the accent is on the last syllable, it acts like an unaccented word except the pitch falls on a particle if one is attached.
Excellent, thank you, if I could give more than a heart I would. Super helpful in more ways than one.
Like, the ways I’ve seen pitch accent explained usually just show how the four patterns work, and that’s fair enough, but something about seeing it explained from the ground up in a slightly more patterned way (eg, always rising on the second mora until the point where they fall, if they do, with the one exception) somehow just makes it click together in my head a lot better. We’ll see how working on hearing it all goes, but something about this alone did a lot to get me closer to understanding rather than, just, here are four arbitrary sound patterns I’ve nebulously remembered.
I had a moment of wondering if particle pitch drops meant that the quiz was expecting me to have memorized how that word works, but I’m guessing what they do is pair that one with choices that you can tell are wrong – basically just not pairing it with the other unaccented form as an option.
Didn’t have a lot to say specifically to others so I didn’t directly respond, but I’m reading it all, and everything has been really helpful.
In tandem with the aforementioned pitch info script, at the WaniKani homepage clicking on your icon in the top-right, then clicking “app” will allow you to enable the audio autoplay option. Give it a dozen levels or so and you’ll be thinking in portals.
If I get a native Japanese speaker to say ame (rain) and ame (candy) consecutively over and over I start to almost here a sinusoidal wave since rain starts high and goes low and candy starts low and goes high…that’s when I hear the pitch differences the most…when placed side by side…it helps me. Don’t know if that’s of any use.
