Sign in

Libre University uses your GitHub account. Signing in is only needed to sit a final test, so the score is kept on your profile.

Quoting and thinking

Every sentence so far has been the speaker's own claim, and a conversation in which nobody can report what anybody else said is a very short one. Japanese does the whole of quotation with a single particle, と, and the striking thing about it is how little it changes the quoted words compared with what English does to them.

と, the quotation particle

Put a clause in the plain form, add と, and follow it with a verb of saying or thinking. The clause is then that verb's content.

田中たなかさんは明日あしたるといました Tanaka-san wa ashita kuru to iimashita Tanaka said he would come tomorrow

The verbs that take と are the ones you would expect: う (say), おもう (think), かんがえる (consider), く (ask, hear), く (write), こたえる (answer), める (decide). All of them work the same way, and と never changes shape.

Two mechanical points before the interesting ones. The clause must be plain, since it is not the final verb of the sentence, which is the rule from the first lesson doing its work again. And nouns and な adjectives take before と, not な: 学生がくせいだとおもいます, しずかだとおもいます. This is worth pausing on, because ので and のに wanted な in exactly the same position two lessons ago. There is no unifying rule to learn; と takes だ, ので and のに and んです take な, and the two sets have to be kept apart by drilling.

Direct and indirect, and how little changes

Japanese has direct quotation, written between 「 and 」, which reproduces the original words exactly, politeness and all.

田中たなかさんは「明日あしたます」といました Tanaka-san wa "ashita kimasu" to iimashita Tanaka said, "I'll come tomorrow"

It also has indirect quotation, which reports the content. Now count what has to change to convert one into the other, in each language.

Converting the English "I'll come tomorrow" into "he said he would come tomorrow" changes three things: the pronoun, from I to he; the tense, from will to would; and, if the report is made on a later day, the adverb, from tomorrow to the next day. Converting the Japanese changes exactly one: ます becomes る, because a non-final verb takes the plain form. There is no pronoun to shift, because Japanese did not have one in the first place. There is no tense shift, for reasons in the next section. The adverb does shift if the day has changed, giving その for 明日あした, but that is a fact about the world rather than about grammar.

The practical result is that Japanese indirect quotation is much closer to the original than English indirect quotation. What it strips is politeness, honorifics and sentence-final particles: 「きますよ」 becomes くと, losing both the ます and the よ, because those belong to the act of speaking rather than to the content.

Example. Convert 山田やまださんは「映画えいがませんでした」といました into indirect quotation.

Only the quoted verb changes, from the polite negative past to the plain negative past: 山田やまださんは映画えいがなかったといました. The brackets go, the ます frame goes, and everything else, including the particle を and the word order, stays where it was. In English the same conversion produces "Yamada said she hadn't seen the film", with a pronoun invented and the tense pushed one step further back; Japanese does neither.

Now you. Convert 先生せんせいは「この問題もんだいむずかしいですね」といました into indirect quotation, and say what is lost.

Answer

先生せんせいはこの問題もんだいむずかしいといました. Two things are lost, and both are things that were addressed to the listener rather than being part of the claim: です, which was politeness, and ね, which invited agreement. The content, that the problem is difficult, survives untouched. A useful check on any indirect quotation you build is that nothing addressed to a listener should remain inside it.

The tense that does not shift

Japanese tense inside a quotation is fixed relative to the moment the words were said or the thought was had, not to the moment of reporting. This is the same relative-tense principle that governed とき three lessons ago, and it is one of the tidiest generalisations in the language.

田中たなかさんは明日あしたるといました means that on the day Tanaka spoke, the coming was still in the future, so it stands in the nonpast, and it stays there no matter how long ago the conversation was. English cannot leave it alone: "he said he will come" and "he said he would come" differ, and choosing between them is a decision about the reporter's standpoint. Japanese never makes that decision, and so never gets it wrong.

The mirror case is equally clean. 田中たなかさんはもうたといました reports that Tanaka, at the time of speaking, said the coming had already happened. Only the final verb, いました, places the whole event in the past relative to now.

Whose thought is it

おもう has a restriction that catches almost every learner, and it is the same restriction you may have met with ほしい at the end of the beginner course. You cannot flatly assert somebody else's inner state. わたしきたいとおもいます is fine. 田中たなかさんはきたいとおもいます is not, because it claims direct access to Tanaka's mind.

The repair is ている, which reframes the thought as an observable ongoing state rather than a claim about the inside of someone's head.

田中たなかさんは日本にほんかえりたいとおもっています Tanaka-san wa nihon ni kaeritai to omotte imasu Tanaka wants to go back to Japan

So おもう for yourself, おもっている for anyone else. The same pattern is why you will see おもっている used of yourself too, where it suggests a settled view held over time rather than an opinion formed on the spot.

One further use is worth flagging now and gets a lesson later: the volitional form plus とおもう, as in こうとおもいます, which expresses an intention rather than an opinion. It is the same と doing the same job, with a different form of the verb in front of it.

って, in speech

In conversation, と is very often replaced by って, which is shorter and completely standard in casual speech. It does the same work.

田中たなかさん、ないって Tanaka-san, konai tte Tanaka says he's not coming

って does three related jobs, and they run together in practice. It quotes, as above. It reports hearsay when the following verb is dropped entirely, which is how the sentence above works: nothing follows って and the listener supplies "he says". And it introduces a topic for definition, standing in for という or とは, as in 納豆なっとうってなんですか, "what's this natto?".

That last job connects to という, which is not casual at all and is used constantly in ordinary speech and writing to attach a name to a thing.

田中たなか」というひとから電話でんわがありました "Tanaka" to iu hito kara denwa ga arimashita there was a call from a person called Tanaka

Note that という is a modifying clause of the kind built four lessons ago, sitting in front of the noun ひと. Nothing new is going on: とう is quoting the name, and the whole thing describes the person.

One more distinction belongs here, because it is the commonest reporting verb of all and learners routinely pick the wrong one. いました reports the act of saying, on one occasion. っていました reports the standing content of what somebody said, and it is what you want when you are relaying a message. 田中たなかさんはないとっていました, "Tanaka said he isn't coming", is the natural way to pass that on; いました draws attention to the moment of speech, which is usually not the point. The same ている that fixed おもう for third persons is doing similar work here.

Example. A caller gave the name Yamada. Report the call to a colleague, and then ask what an unfamiliar word means, using って.

山田やまだ」というかたからお電話でんわがありました, and 「たてまえ」ってなんですか. The first uses という to attach a name to a person you cannot otherwise identify, with かた as the polite word for ひと. The second uses って in its topic-introducing job, which is the ordinary way to ask about a word: you quote it and then ask. Beginners reach for 「たてまえ」の意味いみなんですか, which is correct and sounds like a dictionary.

Now you. Suzuki has told you the meeting has moved to three. Pass that on casually with って, then formally.

Answer

Casually: 鈴木すずきさん、会議かいぎ三時さんじになったって. Formally: 鈴木すずきさんは会議かいぎ三時さんじになったとっていました. Both keep the quoted verb in the plain past, since the change had already happened when Suzuki spoke. The formal version uses っていました rather than いました because you are relaying the content rather than reporting the moment, and dropping the verb entirely after って is what makes the casual version casual.

Embedding a question

To put a question inside a sentence, keep its か and add どうか if the question was yes-or-no.

明日あしたあめるかどうかかりません ashita ame ga furu ka dou ka wakarimasen I don't know whether it will rain tomorrow

どうか is optional and its presence is a matter of emphasis rather than grammar: るかかりません is also correct. With a question word already present, どうか is not used at all.

だれるかっていますか dare ga kuru ka shitte imasu ka do you know who's coming?

なにうかまだめていません nani o kau ka mada kimete imasen I haven't decided what to buy yet

One detail traps people. Before か, a noun or な adjective drops だ altogether: 学生がくせいかりません, not 学生がくせいだか. So the same slot has now demanded three different things, な before ので, だ before と, and nothing before か, which is exactly the kind of thing that has to be drilled rather than reasoned about.

Example. Report that you asked the teacher whether there is an exam tomorrow.

先生せんせい明日あした試験しけんがあるかどうかきました. The embedded question keeps its own か, the yes-or-no shape licenses どうか, and the verb of asking takes に for the person asked, not を. Notice that this is not a と quotation: with an embedded question the か does the work that と does elsewhere, and adding both, as in あるかどうかときました, is wrong.

Now you. Say that you do not know whether Tanaka is a student, and that you have not decided where to go.

Answer

田中たなかさんが学生がくせいかどうかかりません and どこにくかまだめていません. In the first, the noun 学生がくせい sits directly before か with no だ, which is the trap in this section. In the second, the question word どこ is already present, so どうか does not appear, and める takes ている in the negative because "haven't decided" is a state rather than an event.

Vocabulary

WordReadingMeaning
かんがえるkangaeruthink over, consider
こたえるkotaeruanswer
めるkimerudecide
問題もんだいmondaiproblem, question
かるwakaruunderstand, know
というto iucalled, named
ってttethe casual と
かどうかka dou kawhether or not
もうmoualready
まだmadastill, not yet

A clause can now be attributed to somebody. What it cannot yet do is be held at arm's length: everything reported so far has been reported as fact. The next lesson supplies the whole apparatus for saying how far you are standing behind a claim, from a near certainty down to something you overheard.