日本語を学ぼう Learn Japanese

[clawgrip]

Overall, pretty good.

(Lesson 9):

Spoiler:

Exercise:
make these verbs polite:
来る (くる) 来ます kimasu
切る (きる) (u) 切ります kirimasu
言う (いう) 言います iimasu
見る (みる) 見ます mimasu
する します shimasu
話す (はなす) 話します hanashimasu
並ぶ (ならぶ) 並びます narabimasu

Translate the following:
学校で英語を習いました。 Gakkō de eigo o naraimashita. I learned English at school.
郵便局で聞きました。 Yūbinkyoku de kikimashita. I asked at the post office.
学生たちが先生の言うことをちゃんと聞きます。 Gakuseitachi ga sensei no iu koto o chanto kikimasu. The students properly listen to the subject that the teacher is talking about.
池の隣で英語を習います。 Ike no tonari de eigo o naraimasu. I study English next to the pond.

Translate the following, using the polite form:
The students entered the large classroom. 学生たちが大きい教室に入りました。 Gakuseitachi ga ōkii kyōshitsu ni hairimashita.
The students are in the classroom. 学生たちが教室にいます。 Gakuseitachi ga kyōshitsu ni imasu.
The textbooks are in the classroom. 教科書が教室にあります。 Kyōkasho ga kyōshitsu ni arimasu.
The students sat down. 学生たちが座りました。 Gakuseitachi ga suwarimashita.
The teacher taught math to them in the classroom. 先生が数学を教室で学生たちに教えました。 Sensei ga sūgaku o kyōshitsu de gakuseitachi ni oshiemashita.
The students wrote sentences in their textbooks. 学生たちが文章を学生たちの教科書で書きました。 Gakuseitachi ga bunshō o gakuseitachi no kyōkasho de kakimashita.
This small town is old. この小さいが町古いです。 Kono chiisai machi ga furui desu.
as a general rule, you need something to mark machi as the topic. Also this is kind of a weird sentence...what was I thinking when I wrote it? Something other than "town" would have been better.

Spoiler:

It's about half past three in the morning where I am, and unfortunately, I wasn't able to do all of these exercises in one sitting, so I apologize in advance for any stupid mistakes I've made.

Do it at whatever pace suits you.

I'm rather unsure about how I've chosen to translate a few things in general, and about word order when it comes to oblique arguments, but more specifically:

The second and third to last sentences from the third section caught me a little off guard with pronouns/possessive adjectives. I think my translations may be too redundant, especially for "The students wrote sentences in their textbooks".

Right, I shouldn't have included "them" since pronouns are in Lesson 10. A mistake on my part. A good rule of thumb though is that in regular sentences, the direct object comes directly before the verb. Sometimes it gets moved away, so to have it elsewhere is not grammatically incorrect or anything, but it is a more marked form. Just keeping it beside the verb is best for now.

My second to last English sentence also is probably too naturalistic for the sake of the translation. Your sentence is redundant. Actually, no possessive of any kind is required since it's obvious they are writing in their own books.

How compulsory and/or common is "-tachi" for plural animate nouns?

Human nouns are far more likely to take plurals, especially when it is a definite reference.

Hmm… I thought I'd have a greater number of specific questions than that. If I think of more that I may be forgetting at the moment, I'll ask them. Hopefully it doesn't take me as long to start the next lesson as it took me to get around to doing this one.

[zyma]

Spoiler:

This small town is old. この小さいが町古いです。 Kono chiisai machi ga furui desu.
as a general rule, you need something to mark machi as the topic.

Ah, right. Silly me. For some reason, I forgot for a moment there when I was typing that sentence/answer that sentences with verbal adjectives still require subject/object/topic/etc. particles if subjects/objects/topics/etc. aren't dropped/omitted.

A good rule of thumb though is that in regular sentences, the direct object comes directly before the verb. Sometimes it gets moved away, so to have it elsewhere is not grammatically incorrect or anything, but it is a more marked form. Just keeping it beside the verb is best for now.

Oh, OK. Would I be jumping ahead of things by asking specifically what sentences in which the direct object doesn't come right before the verb are marked as? Do different word orders emphasize oblique arguments?

---

Regarding Lesson 10 and pronouns, it surprises me a bit that there are pronouns (that I understand to be at least fairly common) that can be offensive, rude, and/or can make people uncomfortable. I guess that's just because of cultural differences, though. It sounds like context has a lot to do with this kind of thing, and I'm sure it won't be too long before this is no longer surprising to me (I wouldn't be surprised to hear that, at some points, a few speakers of some languages have run into a bit of confusion - at least briefly and at first - while learning English about why you shouldn't really refer to someone as "it"). I'm also rather unsure about what situations would warrant polite/honorific verb forms (and polite/honorific speech in general), but this, along with my pronoun confusion, just goes to show that it's generally pretty hard to study a language without studying the culture that goes along with it. There's no real substitute for immersion and experience and such, and the only time I ever spent in Japan was two weeks when I was about 11, but hopefully I'll be able to find a few reliable/accurate resources about the subject of Japanese culture and how it relates to the language in general to read.

---

(Lesson 11):

Spoiler:

Exercise 1: Translate the following into Japanese:

I do not eat quickly. すぐ食べない。 Sugu tabenai.
The famous actor suddenly died. 急に有名な俳優が死んだ。 Kyū ni yūmei na haiyū ga shinda.
I swam a little and then felt refreshed. 少し泳いですっきりする。 Sukoshi oyoide sukkiri suru.

Exercise 2: Combine these sentences with adverbials
ハンバーグをいっぱい食べた。 + 太った。 ハンバーグをいっぱい食べて太った。 Hanbāgu o ippai tabete futotta.
お母さんはオリジナル商品を作る。自分で売る。 お母さんはオリジナル商品を作って自分で売る。 Okāsan wa orijinaru shōhin o tsukutte jibun de uru.
郵便局に行った。手紙を出した。 郵便局に行って手紙を出した。 Yūbinkyoku ni itte tegami o dashita.

Exercise 3: Translate the sentences in exercise 2 into English.

ハンバーグをいっぱい食べて太った。 Hanbāgu o ippai tabete futotta. I ate hamburgers a lot, so I became fat.
お母さんはオリジナル商品を作って自分で売る。 Okāsan wa orijinaru shōhin o tsukutte jibun de uru. Their mother makes original products and sells them by herself.
郵便局に行って手紙を出した。 Yūbinkyoku ni itte tegami o dashita. I went to the post office and mailed a letter.

Spoiler:

This first question is rather stupid, since it's not very important at all pretty much across the board, and since I have a feeling it might not have a definite answer. Anyway, I noticed that ちゃんと is transliterated as chan to in this lesson, but as chanto in Lesson 9. Additionally, within this lesson, there are five adverbs listed as "Adverbs with to". For four of them, to is transliterated as a separate word, but ダーッと is romanized as dātto (because of the geminate consonant?). Are there any real reasons why some of the things I've mentioned in this paragraph are the way they are? Does the placement of word boundaries ever really matter when writing in rōmaji, or is it somewhat (or maybe even completely) arbitrary, because of how things are written in kanji and kana? Either way, are there any generally accepted conventions for this kind of thing?

Are there any closer equivalents to the conjunctions of many European languages in Japanese than subordinating adverbs, or is context enough 100% of the time? One of the main reasons I ask is that, when I was in Japan/preparing to go to Japan, I learned that と to meant "and". Based on the uses of that word that I can remember hearing, and what you've taught here, I'm starting to assume that it might only be used to coordinate individual words (mostly nouns?) and possibly phrases as well, but not clauses. Is this even close to being correct?

Are there any rules, general or not, about where to place adverbs in terms of word order? I translated "I do not eat quickly" as すぐ食べない。 Sugu tabenai., largely based on example sentences of yours, namely すぐ寝た。 Sugu neta. "I immediately went to sleep" and すぐ家に帰った。 Sugu ie ni kaetta. "I went home right away". For whatever reason, these made me assume adverbs were always clause-initial, despite the presence of example sentences like ハンバーグをいっぱい食べた。 Hanbāgu o ippai tabeta. (possibly meaning) "I ate hamburgers a lot", where the adverb is between the direct object and the verb. Where is the most pragmatically neutral/least marked place in a sentence to have an adverb, if such a place even exists? Can different placements have different connotations, and if so, what connotations can some of those placements have? Hopefully I've expressed this question clearly.

Would it be correct to assume that, generally, the tense of the main verb of a sentence's last clause "applies" to the verbs of previous clauses, since it seems like it would be difficult, if not impossible, to mark a verb (or verbal adjective or copula, of course) as both past tense and a subordinating adverb? If so, could adverbs related to time be used to show that two clauses in the same sentence were not both either past or nonpast, if such a distinction were needed, or would context take care of things at least most of the time?

(Assuming I was at least somewhat correct in connecting two of the sentences provided in the way I did…) In a sentence like お母さんはオリジナル商品を作って自分で売る。 Okāsan wa orijinaru shōhin o tsukutte jibun de uru., would 自分で jibun de only modify 売る uru, or could it be interpreted as modifying the previous clause's main verb, 作る tsukuru, as well?

I'm sorry to say I'm not sure I understood the sections "Two negative adverbs" and "Alternate form of -nakute" very much at all, but I'm not really sure what to even ask about them. Are these endings only applied to "true" verbs, or to verbal adjectives and the copula as well?

---

(Lesson 12):

Spoiler:

Exercise: match the verbs with their meanings, based on the suffix patterns we learned above:

1. 下がる sagàru, 下げる sagèru
go down; descend (vi): 下がる sagaru
lower; move down (vt): 下げる sageru

2. 驚かす odorokàsu, 驚く odoròku
be surprised (vi): 驚く odoroku
surprise (vt): 驚かす odorokasu

3. 出す dàsu, 出る dèru
go out (vi): 出る deru
put out (vt): 出す dasu

4. 変わる kawaru, 変える kaeru
change (vi): 変わる kawaru
change (vt): 変える kaeru

5. 現れる arawarèru,　表す arawàsu
appear; show up (vi): 現れる arawareru
show; express (vt): 表す arawasu

6. 分ける wakèru, 分かる wakàru
be understandable (i.e. be discernible/divisible) (vi): 分かる wakaru
divide; separate; group (vt): 分ける wakeru

7. 染まる somaru, 染める someru
be dyed/coloured (vi): 染まる somaru
dye/colour (vt): 染める someru

8. 渡す watasu, 渡る wataru
cross (vi): 渡る wataru
hand over (i.e. cause to cross) (vt): 渡す watasu

9. 汚す yogosu, 汚れる yogoreru
become dirty (vi): 汚れる yogoreru
make dirty (vt): 汚す yogosu

10. 乾く kawàku,　乾かす kawakàsu
dry (vi): 乾く kawaku
dry (vt): 乾かす kawakasu

11. 伸ばす nobàsu, 伸びる nobìru
stretch (vi): 伸びる nobiru
stretch (vt): 伸ばす nobasu

12. 加える kuwaeru, 加わる kuwawaru
be added; be put in (vi): 加わる kuwawaru
add; put in (vt): 加える kuwaeru

Spoiler:

These were probably the most difficult exercises so far, honestly, and yet I don't really have any questions, other than this:

Would you recommend trying to memorize the more common patterns and exceptions and such, or would you recommend just being aware of them, but still learning verbs themselves as pairs (or even individually, if one happens to come across a verb that has a counterpart which isn't stated there with it)? Hopefully I've asked that clearly.

[clawgrip]

Spoiler:

This small town is old. この小さいが町古いです。 Kono chiisai machi ga furui desu.
as a general rule, you need something to mark machi as the topic.

Ah, right. Silly me. For some reason, I forgot for a moment there when I was typing that sentence/answer that sentences with verbal adjectives still require subject/object/topic/etc. particles if subjects/objects/topics/etc. aren't dropped/omitted.

A good rule of thumb though is that in regular sentences, the direct object comes directly before the verb. Sometimes it gets moved away, so to have it elsewhere is not grammatically incorrect or anything, but it is a more marked form. Just keeping it beside the verb is best for now.

Oh, OK. Would I be jumping ahead of things by asking specifically what sentences in which the direct object doesn't come right before the verb are marked as? Do different word orders emphasize oblique arguments?


It's a little hard to explain and involves things I haven't taught yet, namely the topic. Basically, marking something as the subject rather than the topic can "emphasize" it, that is to say, place focus on the fact that this noun, and not some other noun, performed the verb's action. So, the pattern A=NOM VERB can often imply that "it was A, not B or C, that did VERB." Placing the object before the subject makes this pattern more clear without explicitly marking the object as the topic, a difference that is fairly subtle.

---

Regarding Lesson 10 and pronouns, it surprises me a bit that there are pronouns (that I understand to be at least fairly common) that can be offensive, rude, and/or can make people uncomfortable. I guess that's just because of cultural differences, though. It sounds like context has a lot to do with this kind of thing, and I'm sure it won't be too long before this is no longer surprising to me (I wouldn't be surprised to hear that, at some points, a few speakers of some languages have run into a bit of confusion - at least briefly and at first - while learning English about why you shouldn't really refer to someone as "it"). I'm also rather unsure about what situations would warrant polite/honorific verb forms (and polite/honorific speech in general), but this, along with my pronoun confusion, just goes to show that it's generally pretty hard to study a language without studying the culture that goes along with it. There's no real substitute for immersion and experience and such, and the only time I ever spent in Japan was two weeks when I was about 11, but hopefully I'll be able to find a few reliable/accurate resources about the subject of Japanese culture and how it relates to the language in general to read.

The politeness thing in pronouns is essentially the same concept as the T-V distinction in European languages, just taken to a logical extreme.

I understand how you feel about the politeness thing. I remember early on, I could easily conjugate verbs from plain to polite, and logically I understood that the polite form was more polite, but I couldn't feel it. I didn't have the cultural experience to tell me why or how it was polite. I can say though that plain/polite and honorific/humble language is governed through a combination of social distance/familiarity and social superiority/subordinateness.

---

(Lesson 11):

Spoiler:

Exercise 1: Translate the following into Japanese:

I do not eat quickly. 早く食べない。 Hayaku tabenai. sugu means "immediately; without delay" I recognize that my sentence is somewhat idiomatically ambiguous in English, but I meant the actual speed of eating, how long it takes you to eat. Here, sugu would mean that almost no time passes before you start eating.
The famous actor suddenly died. 急に有名な俳優が死んだ。 Kyū ni yūmei na haiyū ga shinda. I recommend keeping the adverb next to the verb, which is probably its most default location, but anyway, your sentence is not wrong, so I'm leaving it as is
I swam a little and then felt refreshed. 少し泳いですっきりした。 Sukoshi oyoide sukkiri shita. past tense

Exercise 2: Combine these sentences with adverbials
ハンバーグをいっぱい食べた。 + 太った。 ハンバーグをいっぱい食べて太った。 Hanbāgu o ippai tabete futotta.
お母さんはオリジナル商品を作る。自分で売る。 お母さんはオリジナル商品を作って自分で売る。 Okāsan wa orijinaru shōhin o tsukutte jibun de uru.
郵便局に行った。手紙を出した。 郵便局に行って手紙を出した。 Yūbinkyoku ni itte tegami o dashita.

Exercise 3: Translate the sentences in exercise 2 into English.

ハンバーグをいっぱい食べて太った。 Hanbāgu o ippai tabete futotta. I ate hamburgers a lot, so I became fat.
お母さんはオリジナル商品を作って自分で売る。 Okāsan wa orijinaru shōhin o tsukutte jibun de uru. Their mother makes original products and sells them by herself.
郵便局に行って手紙を出した。 Yūbinkyoku ni itte tegami o dashita. I went to the post office and mailed a letter.

Spoiler:

This first question is rather stupid, since it's not very important at all pretty much across the board, and since I have a feeling it might not have a definite answer. Anyway, I noticed that ちゃんと is transliterated as chan to in this lesson, but as chanto in Lesson 9. Additionally, within this lesson, there are five adverbs listed as "Adverbs with to". For four of them, to is transliterated as a separate word, but ダーッと is romanized as dātto (because of the geminate consonant?). Are there any real reasons why some of the things I've mentioned in this paragraph are the way they are? Does the placement of word boundaries ever really matter when writing in rōmaji, or is it somewhat (or maybe even completely) arbitrary, because of how things are written in kanji and kana? Either way, are there any generally accepted conventions for this kind of thing?

Typically, particles are written separately in most Romanizations of Japanese. However, to me, the particle sometimes feels like a part of the word, especially in these sort of adverbs, where the word never appears without the to . If I've written them out as one word, it is because I wasn't paying attention and that just feels most natural. When I split it into chan to it's because I was specifically trying to point out the to. As for dātto, I left it as one word because of the gemination, as you say.

Are there any closer equivalents to the conjunctions of many European languages in Japanese than subordinating adverbs, or is context enough 100% of the time? One of the main reasons I ask is that, when I was in Japan/preparing to go to Japan, I learned that と to meant "and". Based on the uses of that word that I can remember hearing, and what you've taught here, I'm starting to assume that it might only be used to coordinate individual words (mostly nouns?) and possibly phrases as well, but not clauses. Is this even close to being correct?

There are several conjunctions, and I cover some of them in Lesson 15. As for to, it does connect clauses, but in this capacity it is absolutely not equivalent to the coordinating conjunction "and" in English. Again, I cover some of the uses of this particle in Lesson 15.

Are there any rules, general or not, about where to place adverbs in terms of word order? I translated "I do not eat quickly" as すぐ食べない。 Sugu tabenai., largely based on example sentences of yours, namely すぐ寝た。 Sugu neta. "I immediately went to sleep" and すぐ家に帰った。 Sugu ie ni kaetta. "I went home right away". For whatever reason, these made me assume adverbs were always clause-initial, despite the presence of example sentences like ハンバーグをいっぱい食べた。 Hanbāgu o ippai tabeta. (possibly meaning) "I ate hamburgers a lot", where the adverb is between the direct object and the verb. Where is the most pragmatically neutral/least marked place in a sentence to have an adverb, if such a place even exists? Can different placements have different connotations, and if so, what connotations can some of those placements have? Hopefully I've expressed this question clearly.

I would say that a lot of the time, right in front of the verb is natural, but I think certain established OV collocations can push it in front of the object. There are times where the connotation can be altered, e.g. hayaku shukudai o yatte quick-ADV homework ACC do-(IMP) "hurry up and do your homework" is telling the person to stop wasting time and start doing the homework, while shukudai o hayaku yatte means "do your homework quickly," telling the person to complete it within a short time.

Would it be correct to assume that, generally, the tense of the main verb of a sentence's last clause "applies" to the verbs of previous clauses, since it seems like it would be difficult, if not impossible, to mark a verb (or verbal adjective or copula, of course) as both past tense and a subordinating adverb? If so, could adverbs related to time be used to show that two clauses in the same sentence were not both either past or nonpast, if such a distinction were needed, or would context take care of things at least most of the time?

normally the tense of the adverbialized clause is the same as or previous to the main verb. I'm having trouble thinking of an instance in Japanese where a present tense clause would be subordinate to a past tense clause.

(Assuming I was at least somewhat correct in connecting two of the sentences provided in the way I did…) In a sentence like お母さんはオリジナル商品を作って自分で売る。 Okāsan wa orijinaru shōhin o tsukutte jibun de uru., would 自分で jibun de only modify 売る uru, or could it be interpreted as modifying the previous clause's main verb, 作る tsukuru, as well?

this is an instrumental phrase that can only modify the verb of the clause it's in. If you wanted it to modify the first clause, you'd have to include it there as well.

I'm sorry to say I'm not sure I understood the sections "Two negative adverbs" and "Alternate form of -nakute" very much at all, but I'm not really sure what to even ask about them. Are these endings only applied to "true" verbs, or to verbal adjectives and the copula as well?

Yes, I'm sorry to say I kind of dropped the ball there. There are two distinct forms, but -naide only occurs with "true" verbs, as you say. Every time I think I've come up with an explanation, I find a counterexample. I will have to keep investigating, because I feel like there is a clear answer for it, if I can just lay my finger on it.

---

(Lesson 12):

Spoiler:

Exercise: match the verbs with their meanings, based on the suffix patterns we learned above:

1. 下がる sagàru, 下げる sagèru
go down; descend (vi): 下がる sagaru
lower; move down (vt): 下げる sageru

2. 驚かす odorokàsu, 驚く odoròku
be surprised (vi): 驚く odoroku
surprise (vt): 驚かす odorokasu

3. 出す dàsu, 出る dèru
go out (vi): 出る deru
put out (vt): 出す dasu

4. 変わる kawaru, 変える kaeru
change (vi): 変わる kawaru
change (vt): 変える kaeru

5. 現れる arawarèru,　表す arawàsu
appear; show up (vi): 現れる arawareru
show; express (vt): 表す arawasu

6. 分ける wakèru, 分かる wakàru
be understandable (i.e. be discernible/divisible) (vi): 分かる wakaru
divide; separate; group (vt): 分ける wakeru

7. 染まる somaru, 染める someru
be dyed/coloured (vi): 染まる somaru
dye/colour (vt): 染める someru

8. 渡す watasu, 渡る wataru
cross (vi): 渡る wataru
hand over (i.e. cause to cross) (vt): 渡す watasu

9. 汚す yogosu, 汚れる yogoreru
become dirty (vi): 汚れる yogoreru
make dirty (vt): 汚す yogosu

10. 乾く kawàku,　乾かす kawakàsu
dry (vi): 乾く kawaku
dry (vt): 乾かす kawakasu

11. 伸ばす nobàsu, 伸びる nobìru
stretch (vi): 伸びる nobiru
stretch (vt): 伸ばす nobasu

12. 加える kuwaeru, 加わる kuwawaru
be added; be put in (vi): 加わる kuwawaru
add; put in (vt): 加える kuwaeru

These are all correct!

Spoiler:

These were probably the most difficult exercises so far, honestly, and yet I don't really have any questions, other than this:

Would you recommend trying to memorize the more common patterns and exceptions and such, or would you recommend just being aware of them, but still learning verbs themselves as pairs (or even individually, if one happens to come across a verb that has a counterpart which isn't stated there with it)? Hopefully I've asked that clearly.

I would always recommend the practical over the theoretical. Look at the list, be aware of it, but learn actual verbs. The more pairs you learn, the more you will automatically begin to recognize the patterns, and until that time, our good friend context can usually make your meaning clear. My coworker has almost no understanding of the transitive-intransitive thing at all, but she makes herself understood most of the time, because nearly every time you choose the wrong one of the pair, you create a nonsensical sentence that is consequently easily interpretable since swapping out the verb for an extremely similar one suddenly makes an entirely logical statement. Imagine, if I say "Please rise your hand if you have a question," you will not stare at me in bewilderment...you will raise your hand if you have a question. So it is for most of these pairs, even if they sometimes make weird or comical statements, such as giving inanimate objects agency. But oh well!

[Sier]

I hope you're still observing this thread ;) Because I have a question about the /w/ sound. I'm curious how exactly it is pronounced. Here the Japanese /w/ is described as [w͍] (with a left-right arrow below) and it is said that "The Japanese w is not equivalent to a typical IPA [w] since it is pronounced with lip compression rather than rounding". And I'm not really sure how to correctly produce that "lip compression". Can it be compared to a /β/ softly pronounced? Right now, I'm pronouncing it as a very soft /v/, as though I tried to say /w/ in that position.

[qwed117]

I hope you're still observing this thread ;) Because I have a question about the /w/ sound. I'm curious how exactly it is pronounced. Here the Japanese /w/ is described as [w͍] (with a left-right arrow below) and it is said that "The Japanese w is not equivalent to a typical IPA [w] since it is pronounced with lip compression rather than rounding". And I'm not really sure how to correctly produce that "lip compression". Can it be compared to a /β/ softly pronounced? Right now, I'm pronouncing it as a very soft /v/, as though I tried to say /w/ in that position.

Lip compression is when, while rounding, the lips go inward rather than outward. I hope this helps.

[clawgrip]

Actually, rounding is not used to pronounce Japanese /w/, and it should be noted that lip compression does not just mean "no lip protrusion." It still sounds like a regular /w/ kind of a sound, but there is either no rounding or at least less rounding than in English. Check this out:

https://www.youtube.com/watch?v=VpJ7vIuYpxA

Notice that as she pronounces /wa/, she brings her lips close together, but they are not rounded to any significant amount.

[clawgrip]

Lesson 18: Counting (Part 1)
I've had a partial lesson sitting in my drafts folder for a long time, so I decided to finish it up and post it. I hope I didn't leave anything important out!

Counting in Japanese is no simple matter for the learner. To make things easier, I will divide this into a few parts. For now, we will focus on the basic form and ideas.

So lets get into it. First, Japanese employs two distinct sets of numbers: native numbers, and Sino-Japanese numbers. Sino-Japanese numbers are the default counting system, whereas native Japanese numbers only go as far as ten (generally), and are most commonly used only in conjunction with classifiers or other more restricted situations. It's a little complicated, but let's for now just look at the basic set of numbers.

Sino-Japanese (mainly) numbers
These are the normal numbers that are used to count. They can and do appear on their own:

1 一 ichì
2 二 ni
3 三 san
4 四 shì / yòn
5 五 gò
6 六 rokù
7 七 shichì / nàna
8 八 hachì
9 九 kyū̀ / kù
10 十 jū̀
100 百 hyakù
1,000 千 sèn
10,000 万 màn
100,000,000 億 òku

Numbers are combined logically to form larger numbers, e.g.
十二 jūnì 12
二十 nìjū 20
七百三十五 nanàhyaku-sànjū-gò 735

For the mathematically-minded, when a smaller number precedes a larger one, they are multiplied; When a larger one precedes a smaller one, they are added.

Notes:

• "Arabic" numerals are a fair bit more commonly used than Japanese numerals, and for ten and up, Japanese numerals are extremely rare (except to some extent in menus, but I'll get to that another time).
• 万 màn and 億 òku don't normally appear on their own; if no number appears before them, 10,000 and 100,000,000 generally require 一 to appear before them, i.e. 一万 ichìman and 一億 ichìoku;
• 千 sèn occasionally takes 一, but normally doesn't;
• 百 basically never takes 一;
• yòn and nàna are native roots, but they frequently replace shì and shichì;
• kù is an alternate Sino-Japanese form that appears often in quick counting and math-related speech, and in certain set compounds.
• There are a few combinations that undergo sound changes: 三百 sanbyaku, 六百 roppyaku, 八百 happyaku, 一千 issen, 三千 sanzen 八千 hassen. The reason for these changes will be outlined below.

For speakers of European languages, the hardest part about learning Japanese numbers (and really I think numbers in pretty much any Asian language) is that powers of ten are grouped by fours instead of by threes, as is normal in European languages. Consider English:

one
ten
hundred

one thousand
ten thousand
hundred thousand

one million
ten million
hundred million

and so on (unless you're British, I guess)

These are clearly grouped in threes, which is why we put a comma or space or what have you every three digits.
Japanese, however, is much different:

ichì (1)
jū̀ (10)
hyakù (100)
sèn (1,000)

ichìman (10,000)
jū̀man (100,000)
hyakùman (1,000,000)
sènman (10,000,000)

ichìoku (100,000,000)
jū̀oku (1,000,000,000)
hyakùoku (10,000,000,000)
sèn'oku (100,000,000,000)

Obviously, they are in groups of four. The spoken numbers don't match up with the commas at all, but Japanese written numbers follow the European convention anyway. As a result, it's confusing for Japanese speakers, so with larger numbers you will often see combined numbers, e.g. 2万3000 instead of 23,000 or 5億 instead of 500,000,000 to make them easier to read.

There is, sadly, no shortcut to getting this to feel natural beyond repeated use and exposure. I'm terrible with numbers, so it took forever for me to figure these out without doing calculations in my head every time I heard a number.

Native Japanese numbers
These numbers typically do not appear on their own (except the two mentioned above). They are mostly limited to certain compounds. Here they combined with the default counter -tsu

1 一つ hitòtsu
2 二つ futatsù
3 三つ mittsù
4 四つ yottsù
5 五つ itsùtsu
6 六つ muttsù
7 七つ nanàtsu
8 八つ yattsù
9 九つ kokònotsu
10 十 tṑ

Note that tṑ can't appear with -tsu and thus cannot be used in exactly the same way as the others.

So what do these two classes mean exactly? When do we use native numbers instead of Sino-Japanese numbers? To answer that question, we need to learn a little about classifiers.

Classifiers
In the tradition of East Asian languages, counting in Japanese requires the use of counter words, also called classifiers, that are suffixed to numbers. The specific counter word that must be used in a given situation is determined by the specific noun being counted. For those unfamiliar with classifiers, it helps to know that English does something similar (though different) for uncountable/mass nouns.

Consider that we will often say "one sheet of paper" or "one loaf of bread". If you understand this, then you understand the very basic purpose of classifiers, which is to make it clear what exactly is being counted. Japanese is similar, but there are two major exceptions:
1. classifiers are not optional as they are in English (e.g. in English "sheet of" can be dropped, resulting in "one paper", but it cannot in Japanese)
2. classifiers are used even for countable nouns, (e.g. we can say "one dog" with no intervening classifier, but Japanese requires a classifier here, i.e. "one small-animal of dog").

We've already seen one classifier, namely ～つ -tsu. This is a default classifier that usually means one object. It can be used if you don't know the proper classifier to use, though it is never used with living things.

I will cover the grammar of using these with nouns and in sentences more extensively later.

Some Common Classifiers
Let's take a look at a few common classifiers for regular nouns:

個 -ko small objects (also kind of a default classifier if you don't know which one to use)
人 -nin people
本 -hon long, thin objects; trains; bottles; television shows; videos
枚 -mai flat objects
杯 -hai cups, drinks
匹 -hiki small animals
頭 -tō large animals (like say, tigers and up)
羽 -wa birds
台 -dai machines; vehicles
冊 -satsu books
回 -kai times; instances
階 -kai floors/storeys of a building
文字 -moji letters (of the alphabet, etc.); characters (kanji, kana, etc.)
貫 -kan pieces of sushi (sorry, I couldn't help including this one; and yes, it is quite standard to use this)

This is just a sample. There are in fact a huge number of these counters, many of which (over 100) are in common use. I know that at first, it may seem pointless to use these words, but in all honesty, I think that the more you use them, the more intuitive they become.

Combining numbers and classifiers
Classifiers are suffixed to numbers, but there are a number of changes that must be memorized in order to count effectively. They can be summarized briefly as follows:

Generally, Sino-Japanese numbers are used with classifiers, except 4 and 7, which generally use native numbers:
1枚 ichìmai "one sheet"
2枚 nimai "two sheets"
3枚 sanmai "three sheets"
4枚 yònmai ...
5枚 gòmai
6枚 rokùmai
7枚 nànamai
8枚 hachìmai
9枚 kyū̀mai
10枚 jū̀mai
11枚 jū̀ichìmai
12枚 jū̀nimai
13枚 jū̀sanmai
14枚 jū̀yònmai
etc.

yon- sometimes becomes yo-. It is a rare change in absolute terms, but the words it turns up in are very common:
４人 yonin "four people"
４時 yoji "four o'clock"
4年 yonen "four years"

before voiceless stops/affricates and fricatives, the numbers ichì, rokù, hachì, jū̀, and hyaku become iQ-, roQ-, haQ-, juQ- or jiQ- and hyaQ-, where Q indicates gemination of the following consonant (remember that geminate /h/ is [p:]):

～匹 -hiki counter for small animals
1匹 ìppiki "one small animal"
6匹 roppiki "six small animals"
8匹 happiki "eight small animals"
10匹 juppiki, jippiki* "ten small animals"

compare:
２匹 nìhiki "two small animals"

*jiQ- is a somewhat common variant of juQ-. It is more common among older people.

san voices proceeding /h/ to /b/ (or /p/) and irregularly voices some specific classifiers:

3匹 sanbiki "three animals" (-hiki)
３杯 sanbai "three cups/glasses" (-hai)
３発 sànpatsu "three shots" (-hatsu)

The choice between /b/ and /p/ is lexical.

３千 sanzen "three thousand" (-sen)
3階 sankai or sangai "third floor" (-kai)

certain classifiers require native numbers for 1 and 2, rather than Sino-Japanese:
there are a few irregulars (that are important to learn):

～人 -nin is irregular for one and two:
１人 hitòri "one person"
２人 futarì "two people"
but
３人 sannìn "three people"
4人 yonìn "four people"

文字 moji "letter (of writing)" also counter for letters/characters
１文字 hitòmoji
２文字 futàmoji
３文字 sànmoji
etc.

Exercises:
How would you pronounce the following?

３，０００
２億４万
５万４０００

四千
六百三十
二万九千
五億三
一万三千四百十一

１匹
１００発
２文字
１０枚
１人
３本
２０羽

二百個
七頭
一万台
三冊
八百回

The next lesson will be about how to use these numbers to actually count nouns, because it is not always as simple as just putting it beside the noun.

[Lao Kou]

For speakers of European languages, the hardest part about learning Japanese numbers (and really I think numbers in pretty much any Asian language) is that powers of ten are grouped by fours instead of by threes, as is normal in European languages.

There is, sadly, no shortcut to getting this to feel natural beyond repeated use and exposure. I'm terrible with numbers, so it took forever for me to figure these out without doing calculations in my head every time I heard a number.

I feel your pain, bra. Getting sick of doing the calculations one day, in a fit of pique, I finally just seared into my soul that:

hyakùman = 1,000,000

which made further calculations up and down powers of ten faster by far until it became automatic. Later, I added:

jū̀oku = 1,000,000,000

to the repertoire for those chats about the Chinese population. Now I can move about the system pretty quickly.

Classifiers

Consider that we will often say "one sheet of paper" or "one loaf of bread". If you understand this, then you understand the very basic purpose of classifiers, which is to make it clear what exactly is being counted. Japanese is similar, but there are two major exceptions:
1. classifiers are not optional as they are in English (e.g. in English "sheet of" can be dropped, resulting in "one paper", but it cannot in Japanese)
2. classifiers are used even for countable nouns, (e.g. we can say "one dog" with no intervening classifier, but Japanese requires a classifier here, i.e. "one small-animal of dog").

You've mentioned elsewhere that your Japanese informants find Western plurals inconvenient and unnecessary. Well, here's the payback. Who cares what rather vague descriptor of something is? Necessary? Convenient? Hardly. Depends on what you teethe on. That said, I love 'em both.

[clawgrip]

I learned (as in internalized) them a bit differently. I learned 億 first because it's a basic number, rather than a compound one, and I learned those early on. I then learned 十万 because it comes up so frequently when discussing things like monthly pay, rent, international airplane tickets, and so on (in yen). It's to the point where when I talk about yen values in that price range in English, I sometimes accidentally will start out saying a value in the tens rather than hundreds, before realizing I am translating out of Japanese, and there's no way to finish the number in English, like "twenty...I mean two hundred thousand". This mistake of mine only happens when talking about amounts of yen though, unsurprisingly (but interestingly).

Also, the "Japanese informant" in question was my wife.

[Lao Kou]

Since the Chinese language got to me before Japanese did, it makes sense that our experiences would be slightly dissimilar. By the time I did "Lao Kou: The Japan Years", I'd already worked through the numbers.

Also, the "Japanese informant" in question was my wife.

I meant no disrespect to your wife; I believe you've mentioned her elsewhere as finding Western plurals a drag. However, lest I had presumed she was the only one you had talked to who held this view, I hedged my bets with a plural "informants".

[clawgrip]

No disrespect perceived. I don't know if anyone else had the audacity to say it outright, but I know she's said it at least twice, and I'm sure a lot of people think that anyway.

Plurals and counter words have their own uses and are not just the infectious appendixes of the language. One use of counters I like is in the Japanese "game show" (Western genre labels do not work so well for Japanese TV) called Tōsōchū involves players completing missions for money in a story-based scenario while evading "hunters". The hunters are counted with the counter ～体 -tai (body; statue) instead of ～人 -nin (the regular counter for people), implying that the hunters are robots or something not human. It's a nice touch, and is impossible in English.

[Lao Kou]

Plurals and counter words have their own uses and are not just the infectious appendixes of the language. One use of counters I like is in the Japanese "game show" (Western genre labels do not work so well for Japanese TV) called Tōsōchū involves players completing missions for money in a story-based scenario while evading "hunters". The hunters are counted with the counter ～体 -tai (body; statue) instead of ～人 -nin (the regular counter for people), implying that the hunters are robots or something not human. It's a nice touch, and is impossible in English.

I have very fond memories of a show called "常識 ~ 非常識" with celebrities of the day and a set not-unlike the "Match Game"; a lovely game-show-like little Sunday evening bonbon. I do not have fond memories of a Japanese "Candid Camera" version (though American Alan Funt always gave me the creeps anyway); it was positively bizarre and cruel. But we digress...

[Aevas]

I must admit to having only read through the lessons about pitch accent and kanji, so hopefully my questions haven't been answered in some other post! I have really learned from those two posts, though, so big thanks for writing them!

• Are there any straightforward rules for determining/predicting the pitch accent in compounds? Just by looking at some words it's obvious that it changes, but I don't really notice any kind of pattern (although I've only looked at a handful of words).

• The table with all possible on-yomi pronunciations was really helpful! But what about readings ending in -chi, like 一 ichi, 七 shichi, 八 hachi and 日 nichi? Are those perhaps exceptions only present in those words? I hastily scrolled through all kyoiko kanji and couldn't find any more examples of -chi on-yomi readings.

[clawgrip]

• Are there any straightforward rules for determining/predicting the pitch accent in compounds? Just by looking at some words it's obvious that it changes, but I don't really notice any kind of pattern (although I've only looked at a handful of words).

There's no way to predict things 100%, but I suppose in less basic compounds, you can often see the pitch accent of the first element erased and the accent of the second retained.

宇宙 ùchū "(outer) space"
飛行士 hikṑshi "airman; pilot"
宇宙飛行士 uchū-hikṑshi "astronaut"

松 màtsu "pine"
葉 ha "leaf"
杖 tsùe "cane"
松葉杖 matsubatsùe "crutch"

Also, there are a number of single-kanji endings that tend to shift the accent to the syllable immediately before them.

東京 Tōkyō
東京都 Tōkyṑ-to "Tokyo Metropolis"

会議 kàigi "meeting"
会議室 kaigìshitsu "meeting room"

韓国 Kànkoku "South Korea"
韓国人 Kankokùjin "South Korean (person)"

EDIT: (Something I thought was already in this thread but was actually in my Yabushionese thread)

There are many exceptions to the above. Here are some examples:

顔 kao (face)
色 irò (colour)
顔色 kaoiro (complexion)

The accent of the second constituent has been dropped, and it retains the accentlessness first constituent.

英語 Eigo "English"
辞典 jiten "dictionary"
英語辞典 Eigo-jìten "English dictionary"

The accent pattern of the compound does not reflect that of either of its constituents. However, it does match the pattern of the compounds above by accenting the penultimate syllable.

• The table with all possible on-yomi pronunciations was really helpful! But what about readings ending in -chi, like 一 ichi, 七 shichi, 八 hachi and 日 nichi? Are those perhaps exceptions only present in those words? I hastily scrolled through all kyoiko kanji and couldn't find any more examples of -chi on-yomi readings.

I forgot to include those, but they are the only ones I can think of. They are very basic words, of course, and the final -chi seem to act sort of like a variant of -tsu, since it causes gemination just like -tsu, something that chi does not do outside of these four roots.

[Aevas]

Thank you for a thorough reply, clawgrip! So basically, I guess it's better to have a dictionary at hand that marks the accent!

Also, the fact that 今日 kónnichi becomes 今日は konnichiwá, is that just irregular due to its interjectory nature, or does it follow some type of compounding pattern? If I understand it correctly, the normal accentuation pattern would be for は to remain unaccented after an already accented word, right?

[clawgrip]

It's due to it being an interjection, especially one you are likely to say in a loud and friendly way.

[clawgrip]

Lesson 19: Counting (Part 2)
Now that you've learned the numbers and learned how to combine them with counter words, you will learn how to actually count objects. There are two frequently used ways, and I will outline both. But first, some quantifiers:

少し sukòshi - a little; a few; some
いくつか（の） ìkutsuka (no) - some; several; a number of
何<counter>か（の） nan<counter>ka (no) - some; several; a number of
多く（の） ṑku (no) - most; many
いっぱい ippai - many (more conversational)
ほとんど（の） hotòndo no - most; the majority
全て（の） sùbete (no) - all
皆 minnà, minà- everyone
全員（の） zen'in (no) - everyone; all the people involved

There are more, but this is just a selection. These things can stand in place of numbers. What you will perhaps notice is that a number of them have (no) on them. This (no) also occurs for numbers combined with counters as learned in the previous lesson. The no is required when we directly quantify a noun. The quantifier phrase comes before the noun. Some examples:

１枚の切手 ichìmai no kitte - one stamp
２台の車 nìdai no kuruma - two cars
３匹の子豚 saǹbiki no kobuta - three little pigs
４人の歯医者 yonìn no hàisha - four doctors
etc.

These can of course be replaced by quantifiers, such as those from the list above:
何枚かの切手 nànmai ka no kitte - a few stamps
ほとんどの車 hotòndo no kuruma - most of the cars
全ての豚 sùbete no kobuta - all of the little pigs
全員の歯医者 zen'in no hàisha - all of the doctors

However, much more frequently, the quantifier phrase will have its no removed (if it has one) and be placed in the predicate, typically before the verb, with no case marking of any kind. This can be applied both to the subject or object, so:

生徒さんが５人いる。
Sèito-san ga gonìn iru.
There are five students (t)here.

切手を３枚買ってきて。
Kitte o saǹmai katte kìte.
Please go and buy three stamps.

Quantifiers of course can also be used:

リウマチで軟骨がほとんどなくなってる。
Riumachi de nankotsu ga hotòndo naku natte iru.
Most of the cartilage is gone due to arthritis.

リウマチ riumachi - arthritis (rheumatism)
軟骨 nankotsu - cartilage
なくなる naku naru - disappear

The ones that have no removed are those that almost always appear in the predicate, e.g.:

兵隊以外はみんな逃げた。
Heitai ìgai wa minna nìgeta.
Everyone but the soldiers ran away.

兵隊 heitai - soldiers; troops
以外 ìgai - other than

Though the quantifier is usually adjacent to the verb, it will sometimes come before another argument. I feel I am failing a bit here in teaching, because I cannot fully explain what causes this to be more natural in some cases. But anyway, an example:

子供たちが全員プールに入った。
Kodomotachi ga zen'in pū̀ru ni hàitta.
The children all got into the pool.

子供たちがプールに全員入った。
Kodomotachi ga pū̀ru ni zen'in hàitta.
The children all got into the pool.

For anyone confused about this process, it might be helpful to note that a similar, though not identical phenomenon occurs in English, where the quantifier is dislocated from the noun it quantifies:
All the children got in the pool. vs. The children all got in the pool.
Most of the cartilage is gone. vs. The cartilage is mostly gone.

If you want exercises, just let me know.