IxrecOld Japanese

Old Japanese

This page is extremely incomplete and unreliable because I'm not yet capable of reading old Japanese. In fact, this is mostly here for my own sake. Putting together what I know into some guide-like form helps me retain and organize the knowledge better. But since it's not total crap, and you guys might learn something from this, I might as well upload it.

Pretty much all the information here comes directly from the books I've been using to study old Japanese, along with some "supplementary research" (browsing Japanese wikipedia articles) to fill in the details. One of the books is a ridiculously informative dictionary for old Japanese, complete with grammar guide, and the other is an edition of the Tale of Genji which includes the original text, a modern translation and lots of translation notes. Just to be clear, neither book contains any English, which is part of the reason me translating and repackaging the information here might be of some use to someone.

This page is written on the assumption that you have little to no difficulty reading modern Japanese, or "Kinseigo", since I'll be defining things in terms of modern Japanese whenever it's convenient.

Introductory Stuff
Why 中古語 Chuukogo?
What Has and Hasn't Changed
Kana, Particles and Vocabulary
Interesting Vocabulary
Changes to the Kana
Common Particle Meanings
Case Markers and 語尾 Gobi
Grammatical Words
Verbs
Verb Categories
The Jodoushi Model
List of Chuukogo Verb Forms
Example(s)

Introductory Stuff

Why 中古語 Chuukogo?

The history of the Japanese language has four stages:
上代（じょうだい）日本語 - Old Japanese,
中古（ちゅうこ）日本語 - Early Middle Japanese,
中世（ちゅうせい）日本語 - Late Middle Japanese,
and of course 近世（きんせい）日本語 - Modern Japanese.
To avoid any risk of ambiguity, I'll be referring to these as Joudaigo, Chuukogo, Chuuseigo and Kinseigo throughout this page. I'll use the adjectives "ancient" to imply Joudaigo, "old" to imply Chuukogo and "modern" to imply Kinseigo. Also, anything on this page marked with an asterisk (*) is unique to Joudaigo.
Also, up until recently, Japanese had a literary form called 文語（ぶんご） which was separate from the spoken form, or 口語（こうご）.

This page is primarily about Chuukogo, for various reasons. First, 文語 is extremely similar to Chuukogo, so many modern-day Japanese speakers still have some familiarity with features of Chuukogo, which causes them to pop up in otherwise modern Japanese every so often.
Second, Japanese literature took off in the Heian period, when Chuukogo was being used, so most classics of Japanese fiction are written in Chuukogo (like the Tale of Genji) or Chuuseigo (like the Tale of Heike), which is much closer to Chuukogo than the others. For comparison, the most famous works written in Joudaigo are the Kojiki (mythology), Nihon Shoki (history) and Man'youshuu (poetry).
Third, to oversimplify things a bit, Joudaigo relied on a system of phonetic kanji called man'yougana, which was later used to develop the kana we're familiar with (in fact, man'yougana means "the kana used by the man'youshuu"). So even if you're more interested in Joudaigo, it seems like it might be better to learn Chuukogo (or maybe Chinese?) first and then transition to Joudaigo.

What Has and Hasn't Changed

For obvious reasons, most of the sections below will be listing facts unique to Chuukogo. So before all that, I'll give a very rough semi-accurate summary here of what aspects of Japanese seem to have changed between Chuukogo and Kinseigo, and which haven't.

Chuukogo uses the same kana and kanji, but adds two kana (ゐ wi and ゑ we) and introduces several additional cases of alternate kana pronunciations (in Modern Japanese the only case is は being read wa).

The major meanings of the basic particles are the same in both. は and が still mark subjects, を still marks objects, で still marks means/method, と still marks quotes, か still marks questions (though now や does too), the annoying mess of meanings に has is basically the same, etc. In fact, all the particle meanings I would consider difficult are unchanged, so adapting to Chuukogo's particles is relatively easy.
Several particles have additional meanings in Chuukogo. For instance, は and を are now also sentence enders, の can also mark subjects.
There are a few particles which fell out of use after Chuukogo, like にて and やは, but these are outnumbered by more familiar particles.
A lot of the grammatical words in Chuukogo will look familiar, and many will also have similar meanings. A handful will be brand new though (including ones like なる which don't look new at all).

With Chuukogo's particles, grammatical words and verb forms, there is a much stronger tendency for a given meaning to be expressable in several different ways (there are about eleven ways to mark the cause or reason for an event), and for a single particle/word/form to have several different meanings at once (as you can see below, べし has seven meanings). To say it more pessimistically: there's a lot more needless overlap, and a lot more frustrating ambiguity.

A very large chunk of the vocabulary is the same, once you get used to alternate spellings caused by the alternate pronunciations and different conjugation system (e.g. かひなし is the same as 甲斐ない).
Of course, there's also a lot of vocabulary that changed significantly. In my experience, the familiar meanings of these words simply don't make sense in context, which should be enough of a clue to look them up.
Somehow, there seem to be almost no words that have sort of the same meaning but just changed in some subtle way to confuse you (many did change subtly, but so far not in ways that misled me at all).

The way clauses work is, as far as I can tell, essentially the same. The problem is that Chuukogo seems more fond of leaving lots of subjects and objects implicit (apparently referring to people by name was rude in the imperial court, so for all I know this issue might be unique to the Tale of Genji). Also, many of its particles, verb forms and transition words often act as if there's an implicit 事 or 状態 or ように or some other noun or phrase following them. Thus, it's much harder to "guess" how the clauses fit together even though the basic rules are the same. The most common implicit nouns/phrases for a given particle/form/word will be mentioned as part of its definition in the lists below.

The syntactic nature of verb forms is very similar. By that I mean that conjugating still consists of changing the last letter of a verb, attaching a suffix, and possibly repeating with the last letter of that suffix.
However, the majority of the verb forms you're familiar with don't exist in Chuukogo, and most of the verb forms in Chuukogo don't exist in Kinseigo.
One exception is the -i form. Also, compound verbs were just as common back then as they are now.
A partial exception is the -eba form. I say partial because it's no longer a conditional form. Instead of "if _ then _", it now means "because" or "therefore". It's extremely common, so get used to this fast.
As mentioned on the Conjugation page, the jodoushi model of verb conjugation (the one the Japanese themselves use) is much better for explaining Chuukogo than the gaijin model you're used to, so I'll be introducing it below.
Chuukogo adjectives end in し in their default form, not い.

Also, Chuukogo verb forms tend to be noticably shorter on average, which is a major cause of the aforementioned overlap. For instance, the -a nu (negative) and -i nu forms (past tense, sorta) are spelled identically on 一段 and 二段 verbs. It gets worse when these forms are themselves conjugated: an -i te form might be a conjunctive -te, or it could be a form of the -i tsu form (another past tense, sorta). Surprisingly, it's not that hard to deal with this, but doing so often requires looking at what comes before and after each part of the conjugated verb, which is probably not how you're used to identifying forms. For instance, a conjunctive -i te would have no further inflections, whereas the -i tsu form changed to -i te must be followed by something (and yes, it's possible to derive most of that claim directly from the information I give below).

Kana, Particles and Vocabulary

Interesting Vocabulary

Here I'm just going to try listing a bunch of words that either pop up so often you'll just need to know them, or are noteworthy for their contrast with parts of the modern lexicon. This part is very much a mess right now. In particular, I know almost nothing about the pronouns right now.

There are also some words which miraculously manage to have almost exactly the same set of meanings in both old and new Japanese. 色 is one example. I might try listing some of these later.

Changes to the Kana

First, the easy one: The ゐ wi and ゑ we hiragana actually get used, so remember them.

Now, remember how each kana has exactly one pronunciation, with the exception of は being read wa when it's a particle? Chuukogo has a lot more exceptions.
Fortunately, most of the exceptions are reasonably intuitive: はひふへほをづぢゐゑ may be read as わいうえおおずじいえ respectively. I'm pretty sure there are a few others, but that covers the vast majority. As a quick example, the old way of writing 思います is 思ひたまふ, which would be read 思いたまう. Thankfully, conjugation doesn't add any complications to this. For instance, たまふ（たまう） has forms like たまはず（たまわず） and たまへば（たまえば）.

Finally, this isn't really a change to the kana, but it'll probably save you some headaches if I tell you about this now. There are a few kana, like で and だ, which you're used to seeing only as particles or in verb stems. But in Chuukogo, there are dzu-type verbs like 出づ（いづ） and まかづ which can conjugate to things like 出だぬ or まかでたまふ.

Common Particle Meanings

To deal with overlap efficiently, I'm going to use some Japanese terms for particularly common meanings.
1) 感動語尾 - a sentence ending particle used when the speaker is emotionally moved by the action (by the way, I made this one up, it's not a real Japanese term).
2) 逆説 - the word for "contradictory conjunction". In English these include "but", "however", "despite", "although", etc.
3) 原因助詞 - the words for "cause/reason" and "particle". Basically, any particle which (like the modern から) may mark the cause or reason for an action later in the sentence (I also made this one up).

Case Markers and 語尾 Gobi

• は is basically the same, though may be used to add emphasis, or as a 語尾 (without 感動).
• も is basically the same, though it's also a 感動語尾.
• を still marks objects, though sometimes it's a bit closer to と or から than you might be used to. It's also a 感動語尾.
• に has all the same meanings as before, plus it's now a 原因助詞. When used at the end of a clause, it can be a 逆説.
• へ is basically the same
• と is basically the same, though it can also be used to list phrases for a comparison (＿と＿と...)
• な has nothing to do with adjectives anymore, but its negative command meaning still exists, and it's also a 感動語尾.


• の and が have all the same meanings they do in Kinseigo. But they also have each other's meanings (this was a big WTF moment for me). For instance, が can denote possession, and の can mark subjects. There are also some cases where it's easier to interpret an old の or が by assuming an implicit ように, もの or こと comes next.
• から and より also have all the same meanings plus each others'. Of course, they're pretty close in Kinseigo already, so this isn't too bad. They also have a new meaning of "through".


• や and やは mark questions. In addition, the question may be a rhetorical one with an implicit negative answer (＿か？いや、ないだろう）. や can also act either like よ/な, or like やら/など. I don't know how to differentiate between these uses yet.
• か can mark questions or be a 感動語尾. かな is strictly a 感動語尾. かは strictly marks questions, possibly rhetorical ones with negative answers.
• よ is basically the same as its modern usage.
• ぞ is still a 語尾, though it can also be used to provide emphasis earlier in the sentence.
• なり and たり are ordinary 語尾 if the verb is conjugated, otherwise なり has a different meaning (see next section).


• にて can mark place/time (like に/"at"/"on"), means/method (like で), someone's situation/status (like として), or cause/reason (so it's a 原因助詞).
• After a noun or clause, して marks means/method (like で), or a person or group of people who perform an action or are used to perform an action (sometimes overlaps with で).
• After an adjective or a verb in its -i form, して is a 逆説, a 原因助詞, can mean "and", or mark a state/situation.


• *ゆ and *よ can mean "through", "more than", から, or mark a beginning (in space or time). *ゆり is similar.

Grammatical Words

First, the ones that are relatively easy to define.

Don't worry about the little "conjugation" parts for now. They'll be explained a few sections later. Last but not least, here are the tricky ones.

• らむ/らん denotes present actions the speaker knows about secondhand, or is guessing are happening. This may also mark the cause of the action in the next clause. If the speaker does know about it firsthand, this asks why it's happening. Compare to -i けむ/けん below.
• らし indicates an inferred cause of or reason for some event the speaker perceives directly.

  Conjugation:

  ＿

  ＿

  らし

  らし らしき

  らし

  ＿

• なり indicates an event inferred from noises or voices the speaker heard, or something they know through hearsay.

  Conjugation:

  ＿

  なり

  なり

  なる

  なれ

  ＿

• べし indicates likelihood/volition/intent/obvious truth/recommendation/order/possibility. Seriously, it can be any of those.

  Conjugation:

  べから

  べく べかり

  べし

  べき べかる

  べけれ

  ＿

• まじ is the negative of べし, so it indicates unlikelihood/non-volition/non-intent/obvious falsehood/prohibition/negative order/impossibility.

  Conjugation:

  まじから

  まじく まじかり

  まじ

  まじき まじかる

  まじけれ

  ＿

• り denotes actions that are complete, but whose effects persist (like 〜て・ある). Since there are weird restrictions on when this form can and can't be used, the -i たり form (same meaning) is much more popular.

  Conjugation:

  ら

  り

  り

  る

  れ

  れ

べし should look very familiar because its predicative form べき and to a lesser extent its negative form べからず (and べし itself) are still present in Kinseigo. Obviously, it has dropped a lot of those meanings over the past several centuries. I also suspect the modern まい form has a lot to do with まじ.

I don't know for sure how you're supposed to tell the difference between all the meanings of べし and まじ, but I suspect it's very similar to what you do with the -a mu form described below.

Verbs

Verb Categories

Before, I defined "verb type" as the letter a verb ends in by default, which determines how its other forms are spelled. A "verb category" is a set of rules determining these spellings based on type. In Kinseigo, 一段 ichidan and 五段 godan are the only verb categories, and all 一段 ichidan verbs are -ru type anyway, so categories just weren't that important before.

In old Japanese, we have the categories 一段 ichidan, 二段 nidan and 四段 yondan (as far as I can tell 三段 sandan doesn't exist). Technically, 一段 and 二段 are each split into "upper" and "lower" categories depending on whether the stem's last vowel is i or e respectively. This doesn't actually affect how the verbs conjugate, so I will ignore it and write lots of "i/e"s instead.
There are also small categories for irregular verbs. カ変 is short for カ行変格活用 "k-sound irregular conjugation", meaning any irregular verb starting with か, け, き, こ or く (though the only verb actually in this category is く). Likewise, there are サ変, ナ変 amd ラ変 categories. Kinseigo has these categories too, but so few verbs are in them that it's usually not worth mentioning.
Also, adjectives are split into ku-type and shiku-type. In Kinseigo, it's often said there are i-adjectives and na-adjectives, but the vast majority of na-adjectives are really just nouns with a particle after them, so they kinda don't count. In Chuukogo, nari-type and tari-type adjectives are the same way, so I will be ignoring these.

The Jodoushi Model

Now, as promised, I'll tell you the model that the Japanese use to explain Japanese verb conjugation. Long story short: everything that can be conjugated has exactly six basic forms. Depending on the verb/adjective/whatever, some of these forms may have no spellings (i.e. they can't be used at all) or multiple spellings (which are interchangeable), but usually it's one spelling each.

The six forms are:

This particular table is a heavily altered version of one I stole from wikipedia. I bring that up only because, if you want to learn more about the jodoushi model itself, separately from Chuukogo, that link is a good starting point.

List of Chuukogo Verb Forms

Within the jodoushi model, I use "verb form" to refer to certain combinations of one of the six basic forms with either a particle or a jodoushi. Specifically, I mean those combinations where the particle/jodoushi must be attached to a verb (not a noun) conjugated to a specific non-infinitive form (so べし, まじ, etc don't count). That may sound a little confusing and arbitrary, but this matches the intuitive distinction between "grammatical words" and "verb forms" that I've used in the main guide, so it's worth using here too.
Only three of the non-infinitive verb forms actually apply here. The 終止形 doesn't count because it's the infinitive form, and I ruled that out. The 命令形 doesn't count because nothing attaches specifically to it (of course some particles can come after it, like と, but those can come after anything). The 連体形 doesn't count because every particle that attaches to it also attaches to nouns (in fact, when you see a "clause-ending particle" in old Japanese, it's often a good idea to assume an implied noun).
So, using my (purportedly intuitive) definition of "verb form", all complex forms are based on 已然形, 未然形 and 連用形. This should be very comforting, since the 未然形 and 連用形 are analogous to the -te and -i forms you're used to, and the 已然形 only makes two extremely specific and easy to remember forms. Let's get 已然形 out of the way first.

Forms made using the 已然形 Conjunctive Form

Forms made using the 未然形 Incomplete Form

Forms made using the 連用形 Connective Form

Example(s)

Obviously, it would be stupid for me to try and write my own sentences, so I'll be taking a handful of lines I like from Genji or the example sentences in my dictionary. By "expert translation", I mean a translation provided by the book I found the sentence in.
The biggest reason I chose these sentences is that I'm pretty sure I actually understand how all the parts add up to the translation provided. Admittedly, a few things about them still puzzle me, and are thus ignored in my explanations, but considering I've only studied Chuukogo for one or two months total, that's unavoidable.

Old Japanese: かなはざりける命のほどぞ尽きせずうらめしき。
Step-by-step Parsing: The first word is the verb かなふ, which in this case is the same as 叶う（かなう）. It's conjugated to the -a ざる negative form, then the ざる conjugated to its past tense -i けり form, and finally that けり to its predicative form ける because it's being used to describe the noun phrase 命のほど. Thus, かなはざりける is equivalent to 叶わなかった.
For 命のほど, we have to remember that ほど can mean 時間. Essentially, this phrase is meant to refer to the length of a life. The particle ぞ at the end is just emphasis (though it should sound very weird). Next, 尽き and せず are more or less the same as modern Japanese: "exhaust" and "to not". In this case, their combination means something like 限りなく. Finally, うらめし is the same as 恨めしい "reproachful", except it's in its predicative -ki form, because the phrase 尽きせずうらめしき is meant to be describing 命のほど. Notice that the presence of ぞ allows the 尽きせずうらめしき phrase to describe something retroactively without causing confusion.
Putting all that together, we have something like "the length of a life, which is such that (some wish) could not be granted, is infinitely reproachful." In the context of the story, the speaker is bemoaning the fact that his lover's life was too short for her to fulfill a promise they'd made.
Expert Translation to Modern Japanese: (その願いの)かなえられなかった命のはかなさが限りもなく恨めしいのである。

Old Japanese: 秋ならで妻よぶ鹿を聞きしがなをりから声の身にはしむかと
Step-by-step Parsing: First, the -a で form of なる is a negation. Second, the -i しがな form is kinda like -i tai, but with a "were it possible, I would love to _" nuance. From that, you can see how the first part of this sentence means "I would love to hear a deer call for his mate when it was not autumn."
Recall をり means "time". The から here marks a cause rather than a start. Notice the の particle is marking 声 as a subject. The verb しむ is basically 沈む. The かと is the same as modern Japanese. So the second part means "I suspect that voice sinks into my body because of the time", where "time" refers to the season of autumn and "sinks into my body" means exactly what you'd expect.
Expert Translation to Modern Japanese: 秋以外の時に、妻を呼んで鳴く鹿の声を聞きたいものだな。秋という季節だからこのように声が身にしみるかと。

Old Japanese: 右近の司の宿直奏の声聞こゆるは、丑になりぬるなるべし。
Step-by-step Parsing: The first part of this sentence just requires you to know things, so focus on the more interesting second part once you read this. 右近の司 refers to a particular guard whose shift is at 丑 (1-3 AM). 宿直奏 refers to a guard announcing their name when their shift begins. 声聞こゆる is exactly what it looks like: "hearing the voice of (the guard saying his own name)".
Now the fun part. The first なる is the usual verb "to become", conjugated to -i nu form, the nu's predicative form adds a ru, to which the naru particle attaches (inferring event based on something heard, in this case the guard's voice), to which the beshi particle attaches (the subject is third person so interpret it as likelihood). Thus this adds up to "It's probably (based on what he heard) just about 1 AM."
Expert Translation to Modern Japanese: 右近衛府の宿直奏の声が聞こえてくるのは、もう丑の時になってしまっているのであろう。

Old Japanese: さるまじき御（おほむ）ふるまひもうちまじりける。
Step-by-step Parsing: さる is like ある, まじき is the attribute form of まじ. Together these mean something like "_ which should not be" which can be interpreted as "undesireable" or "inappropriate".
御 and ふるまひ are the same as おふるまい despite the weird spelling. も is a particle by itself. うちまじる is まじりあう and -i ける is past tense. So the whole line is "Inappropriate behavior was mixed in", which, based on the context I haven't transcribed, means "Sometimes he acted inappropriately."
The とか in the expert translation probably reflects the fact that ける is for secondhand knowledge.
Expert Translation to Modern Japanese: 時には不都合なおふるまいもなくはないのだったとか。

Old Japanese: おのづからかしこまりもえおかず、心の中（うち）に思ふことを隠しあへずなむ、むつれきこえたまひける。
Step-by-step Parsing: おのづから is like "by onself" in modern JP, here it's like 自然と "naturally". かしこまる here means something like "to be respectfully quiet". There's an え＿[negative], meaning "unable to _", around an おく, which here means "to remain as is". Thus this phrase means "it naturally became such that they were unable to remain respectfully quiet".
心の中に思ふこと means exactly what it looks like: "what they felt in their hearts". 隠す also means "hide" as usual. The tricky part here is that the -i あふ form is using 堪ふ instead of 会ふ, so 隠しあへず means "cannot endure hiding". The なむ is emphatic.
Finally, 睦れ means "intimately" (recall 睦まじい) and this きこゆ is one of the many meaning "say" rather than "hear". The -i たまふ is just formality, and -i ける is past tense. So this part means "they spoke intimately".
Expert Translation to Modern Japanese: お互いに胸の中のことも包みきれず話し合うといった具合に、親しみ申されるのであった。

Old Japanese: 「その片かどなき人はあらむや」とのたまへば、「いとさばかりならむあたりには、誰（たれ）かはすかされ寄りはべらむ。」
Step-by-step Parsing: 片かど is 片才, meaning a meager talent/skill. 片かどなき thus means "without even the slightest talent". The -a mu form indicates likelihood since the subject is third person. Notice that や marks a question here. So this part means "Is it probable for such a talentless/worthless person to exist?"
The のたまふ means おっしゃる. Remember the -e ば is not "if" but "because".
いと means "very". さばかり means それほど. あたり means what you think it does. In this case, the -a mu form is best read as という, and さばかりならむ is referring to the phrase 片かどなき in the previous quote. So the literal is "a (hypothetical) place that is _ to such a great degree". Plug in the reference and this becomes "a (hypothetical) place with such a worthless person".
In context, this is Genji and his caretaker talking about court ladies in the imperial palace, so what this phrase really means is "a place with such a worthless woman".
誰（たれ） still means "who?", the pronunciation is just older. すかす means うまいこといってだます. 寄る still means "to visit", and はべる is simply a formal verb. Once again, the -a mu is likelihood. This adds up to "Who would be deceived into visiting (such a place)?"
Expert Translation to Modern Japanese: 源氏は「一つの取り柄もないという人がいるものだろうか」とおっしゃると、中将は、「まったく、それほどひどい女がいるとしたら、そんな所には、誰がだまされて寄りつきますか。」