So I was reading through A Dictionary of Basic Japanese grammar and I noticed a strange contradiction in the たら and あとで entries.

In the あとで entry we see the following note:

Vinf·past ato de can be replaced by Vinf·past ra when the latter is used with a purely temporal meaning (i.e., ‘after’, ‘when’).

Then it lists following examples

a. ご飯を食べたらすぐ勉強しました。

b. 山田さんはビールを飲んだら寝てしまった。

c. 戦争が終わったら東京にもどりました。

Now when we look at the たら entry:

When S2 in “S1 tara S2” represents a past action, the action cannot be one intentionally taken by the agent after the action or event represented by S1.

When I look back at the examples in the あとで entry, I see that examples b and c make perfect sense and follow this rule correctly. However, sentence a looks to be a volitional act.

I asked a native Japanese speaker and they said that it is correct because すぐ is after たら.

X ご飯を食べたら勉強しました。

I'm quite the beginner so I might be missing something obvious. Can someone explain to me why すぐ makes this sentence valid?

  • 1
    I'm a native Japanese speaker myself, and I quite agree with your friends about すぐ's function, but I can't explain why... Maybe it would be better to memorize たらすぐ "right after" virtually as one new word, independent and different from たら itself.
    – isayamag
    Mar 6, 2015 at 3:15
  • Does your book list 「戦争が終わったら東京に戻りました。」 as a correct example?
    – user4032
    Mar 7, 2015 at 1:03
  • It does. All the examples I wrote were listed as correct.
    – Razzek
    Mar 7, 2015 at 1:27
  • To me, both a. ご飯を食べたらすぐ勉強した and c. 戦争が終わったら東京にもどった are not correct when the subject is the first person and it's a simple temporal clause instead of past habit or imaginary conditional.
    – user4092
    Mar 7, 2015 at 7:53

1 Answer 1


Directly to your question, I don't think the すぐ would make the sentence any more (or less) valid. When it's valid with the word, so is without it, and it's not valid, vice versa.

Your citation from A Dictionary of Basic Japanese grammar

When S2 in “S1 tara S2” represents a past action, the action cannot be one intentionally taken by the agent after the action or event represented by S1.

just doesn't sound right to me, either. The semantics of たら has more to do with the speaker-event relationship, rather than the agent-event one (if I correctly assume the agent is the do-er of an action). That is, when you use たら, you're saying about something unrealized, unexpected, or you have no authorization to interfere (includes habits and general truths) from your own viewpoint.

As the result, たら is quite unlikely to describe what yourself actually did in past. Sentences like


strikes me as if it's in a testimony: "(I saw him) went to study no sooner than finishing the meal."

There are other factors make たら usable or not depending on sentence types (question? order?) or whatever else, but they're too much for this answer.

Lastly, although it may not be a practical advice, I recommend you to refer newer books (yours is from 1989!) for Japanese grammar, as the history of Japanese-as-foreign-language education is relatively short, so it still continues to evolve. Incidentally, my answer bases on a 2008 book, 現代日本語文法 (vol. 6).

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .