After reading all of the replies to this I’m not sute if this adds anything useful to the conversation, but I would interpret this sentence as:
Please bring me two of the cokes that were put in the fridge but aren’t anymore.
Like there were more cokes there, and some had been chilled, and others weren’t, and you explicitly want the ones that had been chilled.
I don’t know if the sentence came with a translation, or if any of the other options would be more natural, as this seems like a unreasonable way to express something that could just be expressed with 冷やしたコーラ and be done with. But exercises might sometimes venture into the unnatural just to offer up usage cases. Especially if there is pressure to make a bunch of them, as I think bunpro tries to do, in order for the exercises to not be answered on memory after a while.
The original translation is “Could I have two of the colas that I have chilled (in advance) in the refrigerator?” (and specifically for the 〜ておく grammar point) so I think 冷やしておいた is the right one since I’m asking for the ones I put in the fridge.
Yeah that works if you interpret it as two separate actions connected by the て form. Feels a bit weird, chilling some coke and putting it in the fridge though…
They don’t overlap in Japanese in terms of meaning, but they can overlap when someone is translating from English to Japanese. I was referring specifically to cases in which a past action is being discussed. There was this example on Japanese Stack Exchange a while back from a man with a Japanese wife. He had apparently asked his wife to send him something. When he asked her if it had been sent, she said 「送っている」. He sat there and continued waiting, only to find out later that it had already arrived in his inbox. The ‘overlap’ exists in such cases, where English would use the present perfect or the preterite. (‘I’ve sent it’, in this example, which could be translated as 送った or 送っている depending on the context.)
My reason for finding 冷やしておいていた is that it made the action seem too far away: why in the world would one say that the drinks ‘had been left to chill’? Why not just say ‘left to chill’? I think that the ‘stative’ meaning works, but it’s just… unnecessary? Just using 冷やしておいた is clear enough, because the bottles were definitely put in the fridge at some earlier point, and there’s no point in the past that’s been mentioned at which that state changed. However, my grammatical reasoning is heavily influenced by my experience with French and the fact that I speak English, so perhaps none of this matters in Japanese.
PS: I’m starting to think that I should try using my Chinese knowledge to figure this out, because the examples in the paper I’m reading now remind me of how the past and present function in Chinese. The problem is that there are definitely differences there too, because Chinese has no conjugation, just additional particles that indicate different facts… I’ve heard people say that Chinese has no tenses, only aspects. I’m not sure I agree, but verbs themselves carry no marks of tense whatsoever. The problem is that Japanese verbs require selection of a verb ending in any case, so there are limits to what I can intuit by analogy using Chinese grammar.
Basically I don’t think it works because I think 〜ておくas an auxilliary is resultative (basically it expresses a change in state due to the completion of an event). If I’m right, the state of being meaning doesn’t make sense because you’re already expressing the change of state (just with a non-stative verb).
Struggling to find a good source on the aspect of ておく though - the closely related てある is definitely resultative and is widely discussed. I’ll leave this as a an overview for that, though I’ve seen the idea in a few places, including some academic papers.
I’m kinda tired at the moment, so I skimmed that page. I agree that てある is ‘resultative’ (which I see used a few times in the article as a synonym of my own usage of ‘stative’, but of course, there’s a difference in nuance). I have to say though, I was surprised when I saw 寝てある. Maggie Sensei clearly says that てある is only meant for transitive verbs. I haven’t gone hunting for other sources, so perhaps she’s the one who’s wrong, but what she said lines up with my experience (and with my friend’s – he’s been learning and speaking Japanese for far longer). The entire thing looks very well researched, and again, I don’t disagree with the idea of てある being resultative – I can’t see what else it might be – but I’m just not certain if everything is accurate, even if it’s written in good faith with plenty of fact-checking. It’s like those academic papers involving double が in a single clause: theoretically possible, and happens here and there in real life, but almost no one does it. Nonetheless, a single error or fringe source doesn’t invalidate the whole thing: I’m just voicing my doubts about a particular point.
Like I said, I was considering the present progressive form. The past progressive form does seem strange. However, if we’re talking about ておく in general, I have a few arguments against the idea that it’s resultative, which would – I quite agree – make the present progressive form unnatural:
“ておいている” brings up tons of results (more than six million) on Google. One of the first is a lesson from Maggie Sensei where she’s explaining the difference between 出したままでいる and 出しておいている in order to indicate that まま doesn’t necessarily indicate intention. This may sound like an appeal to authority or to the wisdom of the masses, but this likely indicates that it doesn’t feel all that unnatural.
Call this a thought experiment, but what sort of mental image do you get when you hear しておく? I’m talking about the plain form, not any derived form. (I honestly believe that it’s the same for any ておく form, including とっておく and 放っておく, so I’d like to encourage you not to dismiss these in any analysis of the structure.) For me, I see something or other being done, then being tucked away into some part of space-time that’s then moved into position in preparation for something else. That’s the action it describes. It’s fairly clear from the way Japanese auxiliary verbs like あげる、くれる、もらう and やる are used that their meanings as auxiliaries are derived from their literal meanings. It’s the same thing for ある and いる: てある is resultative because ある is stative. No other interpretation is possible given the nature of ある. However, in the case of おく, which literally means ‘to put’, it’s fairly clear that the structure ておく (in its original form, not a derived form) can be interpreted in at least two ways: expression of a future action, or expression of a frequent/habitual action. We can also take note of the fact that おく is a point action, which makes any resultative/stative interpretation quite impossible. ‘Putting’ is not resultative in and of itself; ‘putting’ creates a state of having been moved. That’s all. てある、ている and even ておる all rely on verbs that are intrinsically stative in order to get their point across. None of the other auxiliary verbs I’ve just mentioned have such a literal meaning, and so they don’t fit into this pattern.
Now then, to get to the reason I’m tired and what I actually intended to post as a follow-up to what I said earlier… I just went through this paper (not all of it, but whatever seemed relevant and to-the-point) about ‘The Expression of Tense and the Perspective of Situational Cognisance in Japanese’ (translation mine. I’m not sure if it’s accurate.): https://core.ac.uk/download/pdf/147422554.pdf (Note to anyone who wants to read it, which I would encourage if possible: it’s entirely in Japanese, English technical terms and examples of translations aside.)
The paper discusses how Japanese handles tense as compared to English and also examines the importance of the point of view (POV) from which a situation is considered in Japanese. I’m not sure if I’ll be able to capture everything accurately, but it says, essentially, that in Japanese, it’s possible for the POV to change constantly based on context, whereas English has a tendency to require that everything be situated relative to the speaker’s position on the timeline. As such, verb endings like た and る/う simply serve to situate the event being discussed relative to the subject of cognition – that is, the person who is taking the situation in – as opposed to marking points on an absolute timeline like in English. In the section on ている, the author discusses previous conceptions of the differences between ている and た (showing that even native speakers see a certain overlap or link between the two), and mentions two features which I feel are fairly striking:
ている is not necessarily more or less related to the time of occurrence than た, but it does express that the resultant state is still present
ている places emphasis on the resultant state, whereas た places emphasis on what has occurred
Finally, in the (very brief) section discussing いる vs いた, the author says that there’s a certain recognition of the 「いる」state (in other words, that there was an action in progress), and that choosing いた indicates that one envisions the action as finished.
Here are two illustrations from the paper, which I think are fairly instructive:
Almost all the common Japanese ‘tenses’ at once on a timeline
As such, I think we can argue that, according to the author, 「冷やしておいていた…ください」 would be strange mainly because there is no reason to assume that the cola is no longer in the fridge when making a request. There is nothing present in context to justify such a thought. おいている and おいた are both possible (assuming that Maggie Sensei’s Japanese is 標準語), just that the latter would focus on the fact that the cola was put into the fridge at some earlier point, whereas the former would focus on the resultant state (i.e. the fact that they’re still there because they were put inside). As a final point, 冷やしておくコーラをください is possible, but again – particularly with regard to ておく being potentially resultative – we have to consider what it means. In my opinion, based on observations of how the dictionary form is usually used as a modifier, this would mean that the cola is characterised by the fact that it’s been left somewhere to cool, and that this is an intrinsic trait of the cola. Thus, this would imply that such cola can be expected by the speaker to exist, and therefore that it is likely to be habitually present in that state e.g. the speaker knows that his family always keeps a few bottles in the fridge, the speaker is with her friends in a shop that sells chilled drinks etc.
With that, I guess I should go take a rest. I hope this helps anyway who’s confused about Japanese tenses (as I was up to this point, and still am, to an extent), even if it doesn’t settle how 冷やしておく can be conjugated in that particular sentence.
PS: Yes, I realise this is ‘short grammar questions’… I guess that doesn’t preclude ‘long grammar answers’ though…
Fascinating discussion, which I will leave to the experts! I am happy to report that this sentence just came up in my Bunpro reviews again, so I can share the suggested alternative grammar in addition to our main culprit, which for what it’s worth, was the default suggested option. The alternatives are:
I think this (with its derivatives) is the most natural option in the list. I’m not sure why the version containing おいてある isn’t there though, given that the past tense version is, especially since at the point of speaking, that resultant state is still current.
For what it’s worth, 寝てある is lifted from one of the references, and I found a resource in Japanese that says for one particular meaning you can do this…probably rare and subtle enough that just saying you shouldn’t is a better option
Yeah, fair enough, I can’t really defend it and after looking into it more I’m moving away from the idea myself…apparently resultatives are generally intransitive and can’t be used to express the future without more grammar, which clearly doesn’t apply to 〜ておく.
I don’t disagree that you can use it, I’m just not sure you can use it for the “progressive” or “stative” meanings, or that it makes sense in this context.
Those are confusing as hell I think I get it, and what you’re describing is more or less how I think of it…
〜る is for incomplete actions, 〜た is for complete (but not necessarily succesful) actions, 〜ている/た indicates that the action covers a period of time. (using “action” for want of a better word)
Urgh, still not convinced - I can see where you’re coming from if we’re talking about putting things in the fridge, but doing something in advance is the relevant meaning. Trying to reformulate what you’ve said with the “doing in advance” meaning
冷やしておいたコーラ、ください ⇒ The cola that I chilled in advance, please (the action of chilling has happened, the cola might be chilled)
冷やしておいているコーラ、ください ⇒ The cola that I ???, please (I can’t work the second one out…“am chilling in advance” feels a bit simplistic after all this discussion, and especially since the meaning is already covered by おいた, and the “definitely still there” nuance feels better expressed with てある)
I’m really struggling with this construction…my brain keeps interpreting it as “a cola that will chill things in advance, please”
to me the「冷やしておく」implies that the action (chilling in advance) hasn’t happened in the context of the sentence (assuming we’re agreed it’s not resultative). So using it as a relative clause to modify コーラ here feels really strange…
I did a sentence search on ninjal for 冷やしておく+noun and none of the results fit your interpretation (tbf they were all abstract nouns like こと and 効果).
Can you think of a [plain-transitive-verb + noun] construction that works the way you’re describing? I might be misunderstanding what you mean. All the transitive examples I can think of are more like "A for " or "A of " rather than "a noun that is "
Very interesting. Thanks for posting the relevant section. I figured that the sentence had to be from one of the sources cited, but I felt I ought to question it nonetheless. It’s interesting that this use very nearly overlaps with ておく, which is what we’re discussing here.
On the illustrations from the paper, based on what was written in the study
Actually, my understanding of what was said in the paper leads me to interpret them as
る is for actions that are situated after or at the position on the timeline of the person perceiving them
た is for actions that are situated (and completed, as you said) before the position on the timeline of the person perceiving them
ている expresses an action that is closer to someone looking from the た-perspective and extends beyond the る-perspective (notice how the 視点 arrows point to places where the eyeball representing the perceiver was in the previous diagram). In other words, at least in my understanding, it extends into the present even if it started in the past. The authors say that there are two sorts of ている: one for which the POV is situated in the midst of the action, expressing continuous action; and another for which the POV is situated at the conclusion of the action, expressing a state. This makes placing ている in between the two viewpoints described earlier quite appropriate in my opinion.
Comparing ておいている with てある and ておいた
Random little quibble for the fun of it: I think that the alternatives posted by aray indicate that the intended meaning was literal ‘placement’ of the cola. Saying おいてあった would be strange otherwise. Now, more seriously, however, assuming we’re looking at the preparatory nuance (which I do think is a sensible interpretation here, even though I don’t really separate them mentally):
To me, this means ‘the cola that I (or whoever else) chilled in advance and which I still have in that state, please’. The difference between this and 冷やしてある or 冷やしておいてある, in my opinion, is that using ある removes the agent from the equation. With てある, the agent is not important, nor is the action; the resultant state is. It could happen that by some whim, the 氷の女王 decided to send a cool breeze towards those bottles, allowing us to say コーラは冷やしてある. The only thing that てある indicates aside from the resultant state is that an action was done intentionally, according to Maggie Sensei, meaning that the Ice Queen must have wanted to do it for some reason. (That’s something I didn’t know, which I spotted while checking on how to use the structure just now.) With ておいている, the agent remains in the sentence (if only implicitly), the nuance of preparation is preserved, and the state of preparation is implied to be ‘current’. That’s the difference between おいている and てある. Also, as I said in my summary of what the authors expressed in the paper, た places emphasis on the action that occurred, whereas ている emphasises the resultant state. That’s the difference between using おいた and おいている in this context.
'Characterising' a noun with a verb in the plain form
I think your brain is interpreting the cola as the subject of the ‘chilling’ by default. However, interpreting cola as the object isn’t the only element of my interpretation.
This is another possible interpretation provided we consider おく as being in the the ‘future tense’ (which is not distinguished from other tenses in Japanese). It’s not wrong, but that’s not what I was referring to as my interpretation. I’ll try to explain it now.
Certainly, but before I give my example, I’d like to look at what you mentioned as coming from Ninjal, because it’s completely relevant and does in fact match my interpretation:
I don’t have an account and I don’t know which Ninjal corpus you used (it seems there are around ten of them), so I can’t be certain that what I’m about to say is right in context, but こと and 効果 seem like the perfect examples to illustrate my point. 冷やしておくこと is the ‘act of cooling something in advance’, meaning that こと is entirely characterised by that action. Similarly, 〜効果 is the ‘effect of cooling in advance’: once again, the word ‘effect’ is entirely characterised by the action mentioned, and that action is considered regardless of when it happens, and even if it never happens.
Here’s a more concrete example that expresses this idea and demonstrates its independence from reality and time:
「これは人が食べるものなの？」(‘This is something people eat?’) This is a statement that could be uttered in shock upon seeing a particularly repulsive-looking dish, or in indignation in a restaurant that serves horrible food. Here, the idea being presented by the sentence is that of an ‘object-that-people-eat’. The object is intrinsically edible, and habitually eaten. The contradiction with reality that leads to the question being asked is that of the object the speaker sees not falling into the category of ‘objects that people eat’. Similarly, we can say よく切る包丁=‘a kitchen knife that cuts well’ or 父の座る椅子=‘the chair on which Father sits’. In other words, I’m referring to the use of verb phrases that is analogous to that of adjectives, a means of describing an object with a verb that is independent of time. This usage exists for intransitive verbs as well: 町に住んでいる人=‘people who are living in (a particular, usually) town’ versus 町に住む人=‘people who live in a town’ (as opposed to those who live in the countryside). As such, 冷やしておくコーラ would be ‘cola that is chilled in advance’. That’s what I’m getting at. The problem is that English doesn’t allow me to replicate the structure used in Japanese without the passive voice, so it’s hard to convey, even though such verb usage exists in English as well.
I hope these explanations clear things up, even if I can’t guarantee that my way of seeing things is correct.
I don’t think we’re disagreeing, just thinking/expressing it in different ways.
And I think that’s a hell of a curveball for a bog standard ておく(to do something in advance) example sentence (especially given the translation)
I don’t usually either but I forced it since I felt it was confusing the issue…
I’d agree if we were just talking about 冷やしておいた vs 冷やしてある. Can’t say I know what nuance 冷やしておいてある would take exactly (or if anyone uses it), but I think ておく is a far stronger indicator of agency than you’re allowing - especially given that you can’t use it to describe other peoples actions (telling them to do something is not describing for these purposes).
When you put a verb that indicates volition in front of a noun with no explicit subject/object marked, strange things happen especially in the plain form
You’re fixating on the future tense too much, that’s not what I meant. To word it differently, there’s no specific event corresponding to/singled out by 冷やしておく - whether that’s because it hasn’t happened yet, or because we’re talking about the it in the abstract or whatever. A 冷やしておいた might have happened in the past, or be happening as we speak, but if that’s irrelevant to us.
No need for an account: nlt.tsukuba.lagoinst.info/search/
Thanks for the examples, they’ve helped me clear up what was bothering me (a bit).
I disagree with your interpretation of this, and how you use it in that sentence.
If we extend your examples it feels to me like we’re talking about a cola that has the attribute of you often chilling it in preparation for something (…and why you brought in the habitual before has suddenly clicked).
Basically if we reify a “cola” from the class of cola we’re talking about here, then each one of the cola is habitually chilled, rather than there being a class of cola that is as a whole, habitually chilled.
I’m going to avoid this construct like the plague unless I come across a situation where it’s unavoidable and there’s someone around to tell me off.
What I was thinking was more of each cola belonging to a class of cola that is habitually chilled, without necessarily being repeatedly chilled itself (since that would be rather ridiculous). I see it as a characteristic shared by the entire set as a whole as opposed to being a characteristic that can be distributed to each element. The specific situation I was envisioning was something like going to a drink shop whose owner you know well, and where you know there are chilled drinks. It’s natural in that case to refer to the ‘chilled drinks’ as 冷やしておく飲み物. You could also use this structure for making a general statement about chilled drinks e.g. ‘chilled drinks are great in summer, but much less so in winter.’ Just for the sake of providing a few similar examples to assure you that this interpretation does exist, even if I’m going to be forced to cherry-pick because Google isn’t a corpus…
It’s really exactly the same thing as using a ‘that’ phrase to characterise something in English: ‘balls that float on water’, for instance. There’s no need for us to be overly scrupulous about using such a structure, but I think we need to be aware that it tends to be used to refer to an entire category instead of a specific case that’s relevant to a particular situation.
I need help or I will die. I’ve spent an hour searching everywhere and it doesn’t make any sense!!! screams in agony
This is a web comic I read for no reason and tried to translate with adequate precision.
Sentence 「そうそう 最近はスマホってのがあってな？ 」 troubles me a lot. 「～あってな」part to be exact. There is no such grammar anywhere when you can finish a sentence with a て form! How can one even translate that?!
There is. Some sentences even end with ように: I just saw one in the Tobira textbook. What it is is an incomplete sentence whose second half is implied by context. What we need to consider in order to understand this one is the common uses of て. The sort we see the most often as beginners are its use as a verb connector and its use in requests (〜して（ください）). Here, it’s being used as a clause connector, a function which is related to its ability to connect verbs, which allows it to express sequential actions or, in this case (in my opinion, anyway), a sort of reason or justification. The character uses あって instead of ある because he hasn’t finished his thought. He’s basically saying, ‘Oh yeah, now there’s a thing called smartphones… and that makes things super convenient because we can communicate instantly!’ The first speech bubble, which ends with あってな, justifies what he says next.
He’s explaining the smartphone to the cat— something like “You know, these days there’s something called a smartphone.” The て form at the end is very conversational, as though he cut off the following part of the sentence, or as though the next speech bubble completes that thought.
There’s a lot ellipsed in this sentence, which is common in casual speech. A formal written version might be 最近はスマホというものがあります.
I am fighting with “cut off” sentences for a year+ already everywhere and got a lot of experience with them but this time there is な particle right after the て form. If it was 「～あって…」 with triple dots then I would have assumed an omission as usual, but this な ruins that understanding since for me な in this context works strictly as a sentence end particle after which nothing can be. Thanks for the explanation and reassurance, I will add this case to my book…
I understand the confusion, but maybe it would help if you thought of な as a ‘tone particle’ instead of a ‘sentence-ending particle’. If you watch anime, for instance, it’s not rare to hear characters using such particles for emphasis or to give themselves time to think when speaking e.g.
「 私はね！伝説の勇者様みたいな人と結婚したいの！ 」(Rifana-chan from The Rising of the Shield Hero)
Translation if you need it: ‘Well, for me, I wanna get married to someone like the legendary Heroes!’
「ごめんね。あたしさ、人と合わせないと不安っていうか、つい空気読んじゃうっていうか。いや、もう、昔からそうなんだよね。」(Yuigahama Yui from My Youth Romantic Comedy Is Wrong As I Expected aka Oregairu)
Translation if you need it: 'Sorry [seeking agreement]. Me, you know, when I don’t fit in, I’m kinda uneasy, or I end up trying to read the mood or something. Nah, I mean, it’s been like that for a long time, you know? ’
You can watch this analysis from ‘That Japanese Man Yuta’ on YouTube:
(I set the timestamp to the point just before Yui starts speaking.)
Yuta goes through what Yui says and mentions how normal it is for Japanese people not to finish their sentences. You’ll notice she ends a lot of her phrases with particles as she tries to express her thoughts.
I think you can put ね・な almost everywhere in a sentence, to create a little filler/pause.
In fact, I remember a funny video in which an elementary school teacher teaches how to split a sentence into “文節” (basic units of the sentence) and he explains that everywhere you can put a ね , then it’s the end of a 文節.
妹は本を読むのがすきだ -> 妹はね、本をね、読むのがね、すきだね