(Read Description!) Japanese Phonetics Episode 22—LIVE!
Added 2017-07-07 16:28:04 +0000 UTCNote: There are a very small number of 中高 verbs that exhibit their accents on the third mora from the end of the word, rather than the second from the end of the word. This occurs when the second to last mora is a long vowel (ー) or the trailing vowel of a double vowel set, such as the え sounds in both 蘇る（よみがえる）and 考える (かんがえる) (in both of these words the え comes immediately after another vowel, あ). These kind of special sounds force the downstep to shift back one mora, as in yoMIGAeru and kaNGAeru (rule listed in Shinmeikai Nihongi Akusento Jiten second edition, appendix 1, page 55). Apologies for not including this information in the lesson, this is a rather advanced rule which is not commonly covered in pitch-accent resources.
Bibliography
Japanese Phonetics Index Page
Good evening Patrons! The twenty-second episode of Japanese Phonetics is live! In this lesson I address the pitch-accent patterns for the various conjugations of nakadaka verbs.
If you have any questions or concerns leave a comment below. Thank you very much for your continued support, and good luck with your Japanese studies!
Best from Beppu,
Dōgen
Comments
Hi Yurei! In 欲しい the downstep is on the し, as in hoSHIi. たら is often used; I believe I went with ば because I never outright learned the pitch of たら and the resources I used for this lesson didn't list the たら form explicitly, and I always try to avoid teaching anything that isn't in one of the recourses I use, as I don't want to spread misinformation. The NHK book (2016 NHK accent dictionary) I bought after scripting this lesson says that the accent for 惜しい, which follows a similar structure to 欲しい to be oSHIi, and that both Oshikattara and oSHIkattara are acceptable for the たら form. I believe that 欲しい behaves in the same manner, but please take this with a grain of salt as I've always perceived 欲しい to be a bit different than the other い adjectives. Hope this helps, and good luck with your future studies!
Dogen
2020-12-25 06:04:01 +0000 UTCHi dogen sensei. I have a question. What is the pitch accent for (欲しい to want ) , (たら if ) ?. Isn't たらmore common than ば?. 🤔
ʏūʀᴇɪ ᴀᴋᴜᴍᴀ
2020-12-22 19:22:06 +0000 UTCHi Kusterdu, I thought that I had answered this question in a different comment before, but I can't actually find any information on this form in my resources right now. My dictionaries seem to indicate that if there is no attaching particle then 平板 will retain it's normal pattern, so 泣く or naKU becomes naKI, and 泳ぐ or oYOgu becomes oYOgi, but in the resources I'm looking at they are all showing the verbs something after the verb stem, as in naKIte or oYOgite, so I'm not able to give a definite answer at this point. OJAD seems to back up the notion that the verb will retain it's normal pattern, as outlined above, but I'm not 100% confident in OJAD, so I'll unfortunately have to avoid giving a definite answer at this time. Apologies for not being able to help more with this!
Dogen
2020-10-10 01:20:42 +0000 UTCHi Dogen! Are there any accent rules for when a verb is in its connecting form (like 食べ or 切り替え, for example)?
kusterdu
2020-10-09 22:53:20 +0000 UTCHi Paco. Different resources are giving me different information on this topic, but if I'm interpreting the information in front of me correctly, it seems that that 考える is generally said as kaNGAeru, 答える is generally said as either way, with perhaps a slight preference for koTAEru, and that 覚える is always said as oBOEru, as you mentioned. I'm not sure why 覚える doesn't also have two pronunciations, it could be that the was originally two but only one of them became very common, so now most, if not all resources only list one of them. Sorry for not being able to help more with this!
Dogen
2020-09-17 11:42:07 +0000 UTCHello Dogen. On Forvo and OJAD the verbs 考える and 答える appear with 2 accents: kaNGAEru / kaNGAeru and koTAEru / koTAeru. Are they 100% interchangeable in standard japanese? Or is there a tendency towords one alternative? (Does the order in which Strawberrybrown says these words and in which they are listed {kaNGAEru -> kaNGAeru etc.] indicate a tendency?) Last question: Why does 覚える NOT fall into the same category? (as it is always oBOEru) Sorry for bringing up this kind of question again. Greetings. Forvo: https://de.forvo.com/word/%E7%AD%94%E3%81%88%E3%82%8B/#ja OJAD: http://www.gavo.t.u-tokyo.ac.jp/ojad/search/index/sortprefix:accent/narabi1:kata_asc/narabi2:accent_asc/narabi3:mola_asc/yure:visible/curve:invisible/details:invisible/limit:20/word:%E7%AD%94%E3%81%88%E3%82%8B
Paco
2020-09-17 10:19:40 +0000 UTCHi Jan! Unfortunately while I believe what you're saying is accurate, I can't give a definitive 'yes,' as there have been multiple times in the past where I've tried to formulate new pitch-accent rules off of other rules and grammar structures, or let other learners know that the rules they've created are accurate, only to realize later on that this wasn't the case. Thus, while again, what you're saying seems to be true when I think about it right here right now, I'm unable to say, 'yes, this is correct 100%' because this isn't something I can check against all verbs of that type. Apologies for not being able to help more with this! Incidentally, when you say 'ru' form verbs I'm assuming you're talking about about the る verbs that don't gain an extra mora during the aforementioned conjugations. There are many '3 mora nakadaka ru-verbs' which do gain a mora, such as 太る and 走る, and of course, these kinds of verbs conjugate as in fuTOtta, haSHItta, and thus aren't the same as the other type of aforementioned '3 mora nakadaka ru-verbs', and therefore don't exhibit their downsteps on their initial mora. Hope this helps!
Dogen
2020-05-30 11:38:40 +0000 UTCHey Dogen, I have a question regarding the downstep on Nakadaka-verbs in the past- and te-form. You pointed out that the downstep occurs on the third to last mora. Therefore, I was thinking and was wondering about which verbs will have a downstep on the first mora. As Nakadaka verbs need to have at least 3 moras and for u-verbs the last mora gets replaced by two moras those verbs will never have the downstep on the first mora. The only verbs that should have the downstep on the first mora in the past- and te-form are 3 mora nakadaka ru-verbs, like 食べる, 起きる, 掛ける because in the past form the ru only gets replaced with ta or te respectively. Is this correct? Thanks for your guidance, Jan
Jan Westphal
2020-05-30 11:05:10 +0000 UTCHi Enrico. To be frank, I was going to consult with the Shinmeikai and NHK accent dictionaries for these questions, as they both are by far the most in-depth resources I have on the subject (I have another book called アクセントの法則 written by Kubozono, but it's not nearly as easy to sparse), but because it sounds like you've already checked for this information, I'll just give you my gut feelings on these. 1. I don't know the correct answer for this, but I would assume (and believe) that there are multiple correct pronunciations, and that this is a direct result of 'mental spillage' from native speakers who are used to saying 出た with a downstep on the で. That is to say, it may be possible that speakers are so used to saying 出た with a downstep on the で that under certain circumstances it feels somewhat unnatural to shift the downstep further back into the overall compound, rather than leaving it on the で. Or, it may feel odd in general for speakers to shift the downstep back across morpheme boundaries in verbs, outside of the context of 出—I've never seen any literature on this so it's hard to say for certain, but I wager that it could be a contributing factor. 2. I would agree that this sounds like a result of devoicing, and that recordings such as aKETEmita may even be somehow related to this as well. That is to say, and this is no more than a theory, but because speakers are so used to saying ~te-kita, and because this structure is somewhat similar in terms of overall sound to ~te-mita, then some speakers may occasionally 'confuse' the pitch of the two, i.e. use the pitch of TE-kita when actually saying TE-mita—similar to the aforementioned 'mental spillage'. It could also be an issue of emphasis, though that's hard to say without context. I apologize for not being able to help more with these, but I would also like to encourage you to remember that speakers don't always, 100% of the time speak according to the rules (though there is typically some reason, perhaps a non-phonetic reason, for a break in rules). Because both of the issues that you brought up seem to be aberrant pronunciations, as you mentioned, and because you found many more recordings of the 'pronunciation according to the normal rules' then I would chalk it up simply to that—occasional aberrant pronunciations that can't be explained at this particular moment with the set of skills that you, or I, have. Realize that this will occur sometimes, and that more time chasing rogue pronunciations means less time allocated to other tasks, though of course the way you spend your study hours is entirely up to you! Cheers!
Dogen
2020-05-24 11:39:47 +0000 UTCHi Kent! This is a fantastic question. Another example is 考える (kaNGAeru). The reason that the downstep is on the third mora from the end in both this word and in 蘇る is because, according to the 新明解日本語アクセント辞典, the downstep can’t occur on a long vowel, or on the trailing vowel of a double vowel combination—in this case the え before る in both words. Sorry for not including this in the lesson—I had overlooked this particular bit, and will included a small revision to the above post. Cheers!
Dogen
2020-05-24 02:32:40 +0000 UTCI think I found an exception: 蘇る・甦る（よみがえる）. On the dictionary that comes on Mac (スーパー大辞林), there are two listings [3] [4]. Do you know of any other exceptions even if rare? I think this might be because 蘇る was derived by combining 黄泉(よみ) and 帰る(かえる) so the accent stayed on か/が.
Kentaro Kawamura
2020-05-23 22:54:30 +0000 UTCHi Dogen! I have a rather in-depth question, so I apologise in advance for the length of this post. I have some doubts regarding the correct pronunciation of the -te/-ta forms of compound verbs in which the second element of the compound is a 2-mora verb (for example 名乗り出た・願い出た・試みた), and about the -te/-ta forms of expressions like してみた or してきた, which are not technically compound verbs, but they're still highly 'grammaticalised' forms, on a slightly different plane from just two different verbs in sequence. I've found some information here and there, but just titbits, never anything really conclusive or particularly authoritative, so I'd like to lay down what I've learnt so far and ask for your opinion on the matter. I'll split the thing into (1) compound verbs and (2) 'semi-compound' verbs. 1) Compound verbs seem easier, I've always assumed that since they're technically a single verb, the downstep would go on -3, like every other verb, regardless of morphological boundaries. This seems to be confirmed by the NHK dictionary, which lists the pronunciation of 試みた as 'koKOROmita' (the capital letters represent high-pitch, I couldn't come up with anything better), but I couldn't find any specific rules or mentions of this fact, nor could I find any other listings of conjugated forms in either the NHK or Shinmeikai dictionaries. I've listened to several 朗読 recordings of various books, and I'm about 80% convinced that they work just like any other verb, since I've heard pronunciations such as neGAideta and naNORIdeta. But, in one single case, I've also heard naNORIDEta. I've listened to each recording multiple times, even slowing them down, and I'm fairly confident that I didn't mishear this 'strange' pitch accent (though I can't be 100% sure). I can also say that the speaker in question was speaking very correct standard Japanese, and didn't seem to me to show any other 'aberrant' pronunciations, just this one instance in which she said 'naNORIDEta'. 2) The ones I called 'semi-compound' verbs are giving me more trouble. I'd always assumed that, not being compound verbs, they would be simply read in sequence, each verb keeping its own accent, and this certainly seems to be the case for non-conjugated forms such as shiTEMIru or shiTEKUru. There are also many '-te kuru' verbs, such as 取ってくる, which can be pronounced both as compounds, 'toTTE KUru', or separately, 'TOtte kuru', so this particular bit doesn't concern me as much, since I guess that if you use the compound-verb pronunciation, you also use compound-verb rules. But then I found an almost throw-away rule in a book which suggests that '-te kuru' verbs in their -te/-ta forms should be pronounced as -3, regardless of how 'compoundised' they are, so not only do you have 'toTTEkita' and 'toTTEkite', but also 'shiTEkita' and 'shiTEkite' whereas I would be expecting 'shiTEKIta' and 'shiTEKIte'. I assume this might have something to do with de-voicing, and maybe it's not so much an intrinsic property of the '-te kuru' structure so much as the accent just shifting one mora backwards to avoid falling on the de-voiced 'ki'. Listening to 朗読 recordings, I've always found this rule to be correct, that is, I've always heard things like 'shiTEkita', and never 'shiTEKIta'. My main problem lies in '-te mita' forms, however: in this case there's no de-voicing, and there are no 'stable compounds' a la 取ってくる, so I would expect the pronunciation to be regular, as in 'TOtte mita' or 'shiRAbete mita' for 中高 verbs, and 'shiTEMIta', or 'aKETEMIta' for 平板 verbs. And, again, listening to 朗読 recordings this seems to be the case, but in one instance I also heard a speaker say 'aKETEmita'. Again, I listened to the recordings multiple times and am fairly confident I didn't mishear it, and the speaker used completely standard Japanese pitch accent in every other instance. Sorry to bother you with such a long-winded mess, but I would really appreciate your opinion on the proper pronunciation of these cases, which I've found somewhat confusing, and if there are multiple possible pronunciations, on what might be the reason behind that. Thanks, and apologies again (I swear I do not enjoy tormenting people with obscure information)!
Enrico Paolini
2020-05-23 13:40:12 +0000 UTCVery glad to hear that, Makai! Good luck with your continued Japanese studies!
Dogen
2020-04-24 03:56:44 +0000 UTCYes, I do agree, your rule in the 23rd does seem to be easier to remember most of the time. However, regardless of the rule I end up using, thank you for your help! I think, in spending time trying to simplify and check over these rules, I actually ended up learning the original ones pretty well. I usually end up putting the downstep in the right place now without having to recall the rules anyway haha. I'm starting to notice and fix so many verbs I've been pronouncing (and hearing) wrong for so long! I think this series will have a great effect on my Japanese in the long run. And, I appreciate you taking the time to check over this little idea I had.
Makai Moore
2020-04-24 03:11:40 +0000 UTCHi Makai. I had a bit of time and tried to check you hypothesis, and it does seem to check out (minus the uncommon exceptions I mention in this lesson, which also aren't covered by the rules I introduce), but I also have a bit of trouble understanding your wording, and personally find the 'remember the past form of 中高 verbs, as it forms the basis of the て、ても、and たり form' advice in the next lesson easier to remember (at least than the second rule you've put forth), so again, I can't really recommend the rules you've laid out. Thanks for your understanding and good luck with your future studies! Cheers!
Dogen
2020-04-24 02:50:18 +0000 UTCThanks so much for the response. I appreciate how much care you put into not accidentally spreading misinformation. I'm double checking it and it seems to work but I'll be cautious when using it. Thanks again!
Makai Moore
2020-04-23 14:00:29 +0000 UTCHi Makai! Thanks for the message. I unfortunately can't recommend on this method, not because it's not correct, but because there have been multiple times in the past in which I thought one of my student's unique summary of certain rules in a lesson made sense, only to realize after the fact, often times several weeks later, than even though said summary worked for certain rules, it often ended up complicating things in the long run, or didn't adhere to certain exceptions. Thus, I would suggest that you follow the information presented in this lesson, though if you're 100% convinced that this rule will make things easier for you I think it would be OK—provided you double check each point. I'd like to also mention that I actually present a 'summary' rule of my own that can help with some, albeit not all, of the conjugations you've mentioned, in the next lesson (lesson 23, from around the 6:10 mark). Sorry for not being more help here, and let me know if there is anything else I can try to help you with. Cheers!
Dogen
2020-04-23 03:49:50 +0000 UTCHello Dogen, great video! I love how you always try to simplify these rules into an easier to digest manner for us. Though, I think I may have found a way to simplify the three 中高 common conjugation groups (た・て・れば, ろ, ても・たり) and I wanted to make sure I wasn't messing something up. If you instead define the downstep based on how many mora it is away from the conjugation ending, I found it creates only two groups. First, you would have the r group (ろ, れば) in which the downstep occurs on the last mora before the conjugation ending. For example, with 走る → 走れば, し is the last mora before the れば so you know the downstep is on the し. The second would be the t group (た, て, ても, たり) in which the downstep occurs on the second to last mora before the ending. For 走る → 走っても, し is the second to last mora before the ても ending so you know, again, the downstep is on the し. On the other hand, with 食べる → 食べても, た would be the second to last mora before the ても hence the downstep being on た. I wanted to clarify if this method works since it would probably make memorizing this section easier, for me at least. Thank you so much for these videos and I hope you’re doing well during these times.
Makai Moore
2020-04-22 18:00:06 +0000 UTCHi Charlie! Yes that is correct—great question by the way. Native speakers usually put the downstep on the け due to devoicing, as you mentioned. Cheers!
Dogen
2019-09-16 03:49:25 +0000 UTCHi Dogen, For verbs like 助ける conjugating into its ～て form, since the verb is [3], would the downstep occur on the す? I have listened to some Forvo recordings of 助けて and it seems to be read as [3]. I have a feeling it is because of the devoicing of す, but I wanted to make sure I wasn't assuming anything.
Quiche
2019-09-16 00:28:01 +0000 UTCHi Jokin, you can learn more about these in the 新明解日本語アクセント辞典、cheers!
Dogen
2019-05-10 00:01:16 +0000 UTCHi Dogen, Just out of curiosity, where I can find more about those exceptions you talk about in the common conjugations of 中高？ Thank you!
Jokin Pedreño Alcolea
2019-05-09 22:45:43 +0000 UTCHi Walt. Apologies for the confusion. By this I did mean the dictionary form.
Dogen
2018-09-17 08:09:51 +0000 UTCExactly what do you mean by "non-conjugated forms"? Do you mean any verb form that can be modified with things like ます、 ました、 and ません? Note: this may actually be part of the question from Dolores (except that her Japanese is way ahead of mine, so she was able to be more specific). Or do you just mean what I know of as the dictionary form?
Walt Poor
2018-09-13 01:19:47 +0000 UTCHi Dolores. In both of those words the downstep occurs on the べ. Though with 食べなくても people sometimes put it on the な as well. Hope this helps ^^
Dogen
2018-02-22 02:18:36 +0000 UTCHi Dogen! Do these rules apply to the negative versions as well? For example, would the down step for 食べなくても and 食べなければ be on the na?
Dolores Derksen
2018-02-21 02:38:58 +0000 UTCHi Enrico, Many words have multiple correct pitch accent patterns, for a variety of different reasons (phonetic shift over time, dialect normalization, etc.). The important thing to know is that you can say these words with either of the pitch accent listings and either of them will be correct. Cheers, Dogen
Dogen
2017-09-28 14:43:40 +0000 UTCHi Dogen! It's me (surprise), sorry to bother you yet again. In this lesson you mention that all 中高 verbs have the accent fall on the second-to-last syllable, and while that is definitely the case for every verb I've seen so far, I have also seen some verbs (for example 振り回す) where there's a secondary accent listed which is still 中高, but with the accent finally on a different syllable. In the case of 振り回す, the NHK dictionary lists the primary pronunciation as per your rule, but the secondary pronunciation as furiMAwasu. I was wondering if there's a particular reason for this, if maybe the secondary pronunciation is a slightly less accepted variation, some sort of corruption or language shift, or whatnot. Thanks again! =)
Enrico Paolini
2017-09-28 14:28:12 +0000 UTCI usually suggest watching through 1-2 times with subtitles such that you can pick up on meaning, and then turning them off after that such that you can concentrate on phonetics :)
Dogen
2017-07-19 23:58:50 +0000 UTCHi Rooky, I've never really thought about a song approach before, but if it works for you go for it! I personally find that repetition is the best for memorizing words which don't fall into any of the easy to remember groups. For example, I usually pronounce a word like 言葉（は）10-20 times in a row, then move onto a new word. Recording yourself always helps tremendously as it forces you to become very aware, which helps with the memorizing process.
Dogen
2017-07-19 23:58:02 +0000 UTCAlso, I've been rewatching "Bakuman" every day at least once and it's making my listening improve at a rapid rate! Also, I find myself translating whole monologues and hearing phrases at each new pass that I hadn't heard previously. (I'm doing this through Japanese Netflix. My question is: You only suggest watching things without subtitles, right? Because that's what I'm doing.
Rooky
2017-07-19 20:21:38 +0000 UTCDogen-sama! Is the best way to memorize your 1-4 mora noun patterns to to do it like mini-songs? The way we learned our ABC's song? So that when the word comes up we'll mentally hear it and know which pitch accent it is? Or would you suggest memorizing by saying "あした is Odaka, 言葉は is odaka, etc Thanks!
Rooky
2017-07-19 20:19:49 +0000 UTCGreat question Dominik. I've found that 関西人 'speaking' 標準語 often follow 標準語 grammar and semantics, but retain a subdued 関西弁 pitch-accent. In other words, the pitch-accent usually isn't 100% 標準語, even if everything else is. Just my own personal experience, but I've seen this quite a bit. Thus, I would personally recommend trying to befriend at least a few people from 関東, especially when it comes to language partners.
Dogen
2017-07-19 14:49:38 +0000 UTCHi Dogen. Coincidentally, most of my current language exchange partners are from Kansai, but they usually speak 標準語 to me. Does that mean they also adjust their intonation patterns or do they still pronounce everything with a Kansai accent? In the latter case, would you recommend me to ditch them and get some friends from Tokyo （∵）？
Dominik
2017-07-19 08:58:19 +0000 UTCHi Patrick, I don't know any beyond the rules that I spoke of in the previous lesson. I've thought about this myself, and tried to analyze verbs based on this, but have never been able to find a pattern that would be beneficial for trying to remember pitch accent. If I ever come across one, however, I will be sure to let you know!
Dogen
2017-07-11 10:52:25 +0000 UTCHi Dogen. Are there any rules (or even just partial rules) relating the pitch accent (nakadaka, atamadaka, heiban) of a verb and its grammatical class (ichidan, godan)?
Patrick R
2017-07-11 01:40:54 +0000 UTC
