IceCream wrote:

In my mind, there is only one grammatical catergory for passive, and the intransitives are included in it.

This is wrong, but the difference between the active->passive and the transitive->intransitive changes is fairly subtle.  They do share an important point, which I'll get to below.

In this case, the eating (of the insect) was done by the fish on the insect. So, the full setence would be 虫が魚に虫を食べられた。The ambiguity arises only from thinking in the wrong amount of categories.

Exactly right.

Now, like you said, it's important to define your terms before you start.  I'm going to examine the action covered by the verb 落とす／落ちる／落とされる／落ちられる.  There are four roles I'll examine:

Agent: causes the event to happen.
Patient: immediately affected by the action.
Passive Patient: affected by the consequences of the action.  As Tae Kim says, the passive patient "has to take it like a bitch."
Destination: the location the patient falls/is dropped.

I'll start from 落とす.  This verb assigns the roles to particles as follows:
Agent -> が
Patient -> を
Passive Patient -> UNASSIGNED
Destination -> に

落とした　(dropped it)
蛇を落とした  (dropped a snake)
友達が落とした (friend is the one who dropped it)
友達は蛇を落とした (friend dropped a snake)
蛇は彼女が箱に落とした (the snakee was dropped by that girl into a box)

English uses passives primarily to move topic-focus around a sentence.  は is capable of that function without the passive mood.

Now, let's switch from transitive to intransitive.  The difference is in the role assignments:
Agent -> UNASSIGNED
Patient -> が
Passive Patient -> UNASSIGNED
Destination -> に

Now that the agent role is unassigned, this verb describes an action that happens without someone causing it.
落ちた。　(it fell)
蛇が落ちた。(snake is thing that fell)
蛇は落ちた。(snake fell)
蛇は箱に落ちた。(snake fell into box)

Compare the switch from active transitive to passive transitive:
Agent -> に (first)
Patient -> を
Passive Patient -> が
Destination -> に (second)
The agent is not missing, just moved off of the が particle.  Also, the Passive Patient role enters play.  The overall effect is to express the effects of the event (direct and indirect) while allowing the agent to be present, but in the background, scheming evilly.
落ちされた。 (it got dropped)
蛇が落ちされた。(snake was thing that got dropped)
蛇は落ちされた。(snake got dropped)
蛇は友達に落ちされた。(snake got dropped by friend)
蛇は友達に箱に落ちされた。(snake got dropped by friend into box)
So far, the Passive Patient seems to be equivalent to the Patient role we've seen earlier.  That's because the result stops there.  If that role is left unfilled, though, we get an expression of second-order effect.
蛇を落ちされた。(someone got snake dropped)
私は蛇を落ちされた。(I got (my) snake dropped)
友達はあの奴に蛇を落ちされた。(friend got (his) snake dropped by that guy)
蛇は友達があの奴に落ちされた。(the snake was dropped by that guy, impacting (emotionally) friend) (Contrived Example Warning!  See below.)
蛇は友達にあの奴が落とした。 (the snake was dropped by that guy onto friend) *not passive mood*

Now, the point I'm not sure on is this: if a noun is assigned to the Passive Patient role with が, can another noun be assigned to the Patient role with は or を?  I've seen PPはPを, but are PPがPを and PはPPが as well?  Do they make sense?  Does anyone ever speak like that?  Certainly something to look for in my reading, but right now I'm not sure.

Finally, we have both at once, the passive intransitive.  Roles:
Agent --> UNASSIGNED
Patient --> に (first)
Passive Patient --> が
Destination --> に (second)
Now this one's weird.  It let's us blame things that "just happen" (intransitive) for affecting someone or something else (passive).  The Patient role picks up a very agent-like sense, even though literally it isn't responsible.
落ちられた。(got fallen)
蛇に落ちられた。(got fallen by snake)
私は蛇に落ちられた。(I was affected by a falling snake.)  Note: does not say that "I" am the destination of the fallen snake, only implies it.
友達が落ちられる。(friend is the one falling affects)
箱には蛇に落ちられた。(Box was fallen into by snake.)

Clear?  Maybe?  Like I said, there's a similarity between passive transitive and active intransitive.  Now it should be clearer: look at the role of が: it's the Patient in the first case and Passive Patient in the second.  The difference is what the verb says about the Agent.  Intransitive verbs have no Agent, passives do but might not specify it.

Oops, I've missed out on another use of the passive, as in: 云われます. In this case, the point is to make it clear that you're not specifying the agent.  「赤い車は一番速いと云います」could be "I/he/you say red cars are the fastest," but 「赤い車は一番速いと云われます」is "people say that red cars are the fastest."

(BTW, old Kanji; if you don't recognize it, shame on you for not reading more 宮沢賢治)

Next time: The verbs 落とさせる、落ちさせる、落とさせられる、落ちさせられる: The Causative Agent.

This thread could scare anyone away from Japanese.

I didn't mean to say that you couldn't have に clauses attached to passive verbs which did not indicate the agent, only that if the agent appears then it must be に marked. You'll note that generally in the examples you've quoted the agent doesn't appear at all.

鈴木先生は日本へ帰られた。

This is the keigo which uses the same verb forms as the passive. It's not a 'real' passive sentence (and you can tell because the grammar works exactly the same as if it had been 帰った, ie the person doing the action is not に-marked). (The 林先生 example later is in the same category.)

the patient can take は instead of が？

I tend to think of it the other way round -- the patient takes が (three times daily, with food). As usual you can replace it with は (or omit it and let it be understood from the は-marked topic, however you like to think of は-が) in the same way as you usually might.
There's nothing magic about は/が in passive sentences.

* この展示会はある著名な日本の建築家の作品を中心に、東京に建てられたモダンな建築について紹介している。
This exhibition, focusing mainly on works by an eminent Japanese architect, introduces modern architecture in Tokyo.
P - ある著名な日本の建築家の作品
D - 東京
MPV - 建てられた

Misparse -- ...の作品 attaches to を中心に (which has an implicit して I think). The P here is the head word of the relative clause, ie モダンな建築.

wildweathel wrote:

この本は１９６５年にアメリカで出版された。
This book was published in America in 1965.

You could substitute 「した」with the same meaning, this passive is purely for feel (more polite, refined, whatever)

Really? Wouldn't ...出版した mean "I/We [or other implicit subject] published this book in America in 1965" ?

林先生は日本の大学のことを話された。
Prof. Hayashi talked about Japanese universities.

調査の結果、新しい事実が発見された。
As a result of the investigation, new facts emerged.

Same-same.  Academic style.  Linguistically boring unless you want to imitate it.

I don't think you could say 新しい事実が発見した at all -- that feels wrong to me. を if you're going to use active here.

I'm confused about why you lumped the 林先生 example in with the other two because to me it is qualitatively different.

the only question is how the ambiguity arises, and how to define it

Are we still talking about Thora's 渡された example here? The cause of the ambiguity seems straightforward enough to me. 渡す allows you to specify the role 'person object is given to', marked with に. In general moving a verb from plain to passive shuffles some of the roles around, but roles not explicitly affected by passivisation get left alone (so for instance if the verb let you specify location with で so does the passive). So 渡された inherits this, and also lets you specify 'person object is given to' with に.

The ambiguity comes because with any verb in the passive form you can also use に to specify the agent  -- so if a verb like this has a に clause it is either (a) unclear which was meant or (b) only deducible from context. Most verbs either don't take に or else に doesn't mark an animate actor, so it's pretty easy to tell which is meant.  (For instance, two examples of different uses of に with the passive of 落す:  米軍に落された民間機 "the civilian plane taken down by the American military" : に indicating agent, vs 長崎に落された原爆 "the atomic bomb dropped on Nagasaki" : に indicating target. [PS: I hope my examples don't offend anybody (or derail the thread) -- selected only because I felt they were the clearest ones google had to offer for the verb.])

[short summary: 渡す is a slightly unusual case and you should be wary of building your understanding of passives on what it does rather than on what a broader range of verbs do.]

Last edited by pm215 (2009 December 22, 3:13 pm)

IceCream wrote:

i've been looking for an example, but i really can't find any where the が works with a transitive passive + an agent, in this way. I think it can't... because it doesn't make good sense that it would. The が does intransify (sort of) by nature, it seems...

google wrote:

彼が彼女にふられたみたいです
日露戦争さなかの1905年早々、旅順要塞が日本軍に落された。

Transitive passive verb, agent marked by に, patient marked by が. Or did you mean something else?

mmm, sorry, i know this isn't helping you, and, in the end, yeah, this kind of thing isn't so important for understanding, but on the other hand... its kinda fun, and i'm actually learning a ton

Oh, I'm having fun too :-)

IceCream wrote:

Even if you don't accept that 米軍に落された民間機 IS 民間機（が）米軍に（民間機を）落された, you still have to accept that putting in the 民間機を　doesn't change the meaning / agents / whatever you want to call it of the sentence. 民間機 still gets the action.

I don't have to accept that :-) My take on this is that adding the を clause (a) creates a rather weird sentence (b) creates a sentence with a different grammar structure. 民間機が goes from being "immediately affected patient of the action" to "indirectly (emotionally) affected patient of the action". Just because you've managed to fill the roles with duplicates doesn't make the two structures grammatically identical. Otherwise "the dog ate himself" and "the dog was eaten by himself" would be the same grammar structure, which they clearly aren't.

However, you can't put an を in 原爆が長崎に落された. without changing what "gets" the action. If you put an を in, it would change the が to an を, and the bomb would no longer be the thing "getting" the action, it would probably be 長崎.

The only difference here is that the set of things already specified prevents you from playing your trick of filling everything with duplicates so we don't notice you changing the grammar structure :-)

pm215 wrote:

IceCream wrote:

In fact, in the japanese equivalent sentence, the guy is the passive patient to begin with, because it's going to be assumed that "he got the dumping and doesn't like it", because he's the は・が marked person. The only reason he seems to be the normal patient in that sentence is because he is actually both the patient and the passive patient at the same time.

I don't think he's the passive patient. That is, I don't think that just being the subject of a plain passive implies any emotional effect on the subject in that way that it does for an 'affective passive'[*].

The difference between the Passive Patient, and Patient (as I use the terms) is not that the Passive Patient is the recipient of an affective effect and the Patient the recipient of a direct effect.

The difference arises in that a passive has two parts: the main verb and the passive inflection, both of which have a patient.  Passive Patient receives the whole event--possibly but not necessarily in a metaphorical sense.  Patient elaborates the action, but receives just the action, not the event.

Consider the verb 受ける.  This is a patient-subject, theme-object verb meaning "to accept, receive, take."  It's not actually used as an auxiliary verb in natural Japanese, but imagine if it was:

殺すのを受けた
Took the event of someone killing someone.

That means something similar to the passive.  We have two patient positions available: the patient of 殺す and the patient of 受ける.  Consider the difference between:

ケニーを殺すことを受けた
Took the event of someone killing Kenny.

殺すことをケニーが受けた
Kenny took the event of someone killing someone.

If we make them passives, they become:

ケニーを殺された　took event of Kenny being killed
(Kenny = Patient of 殺す = Patient)

ケニーが殺された　Kenny took event of someone being killed
(Kenny = Patient of 受ける = Passive Patient)

(The running gag in サウスパーク is 「何だ!?ケニーが殺されちゃった！／この人でなし！」)

There seems to be a convention that が automatically fills both roles, meaning that if が is used with a passive verb, you can't have a separate Patient marked with を or は.  But, that may just be a result of my limited exposure.  All the times I've seen a passive where the two roles are assigned to different nouns, the Passive Patient has been the topic and thus gotten は.  Whether this is due to high correlation between that construction and topic-focus on who receives the event or whether 「…が…を…られる」is impossible I don't really know.

IceCream wrote:

日本が米国に民間機を落とされた　means that Japan got the action of the civilian aircraft being brought down by america, (and isn't happy), or, in your words, 日本 received the whole event. Using the は would just place the emphasis on it being Japan that got the action and not someone else.

I'm not sure if that sentence is actually something you could say, since it uses both が and を with a passive.  I'll keep my eyes open.

On your second point, I feel we're talking past each other (argh! always frustrating).  As far as I know, there's nothing in the grammar that says the PP suffers, or even has an emotional reaction to the event: it's just that the "result stops there" so to say. 

携帯が壊された has one missing role: A.  Someone broke the cell (the passive voice says that if the speaker wants to identify that person, they'll use に, so don't look at the topic)

携帯を壊された is more unfinished: the PP is missing, too.  Someone/thing did the action of breaking cell to someone/thing--and the listener is encouraged to fill it in from context/topic.

I finally finished reading all the posts in this thread, and, well, where should I begin? Hmm... I guess I'll begin with the very basics of the られる thing...

There are multiple grammatical categories when it comes れる/られる such as 受け身 (passive), 可能 (potential), 尊敬 (polite), etc. But the very center of all the れる/られる meanings is 自発. It doesn't seem your average textbook for foreigners teaches this, but it's the most important grammar point of れる/られる because etymologically every other category including passive is derived from this 自発 meaning. Actually passive, potential, etc. are all loosely connected with each other, and there are many examples where れる/られる sits on the border of multiple categories such that you can't determine if it's passive, 自発 or other usages.

So to answer to the OP, it's totally ok if you can't tell whether it's passive or potential in some sentences. When you hear/see a sentence, the intended meaning is there, right in front of you. You don't need to analyze it to label a word as long as you understand what it says. Human languages are all like that. Just because it's a foreign language doesn't mean it's less ambiguous or more logical than your mother tongue.

I guess some posters are arguing about special cases where れる/られる is clearly passive, so I'll explain Japanese passive voice a bit.

There are two main categories of passive: Direct (直接受身) and indirect (間接受身). The former only takes so-called "transitive" verbs while the latter can take both transitive and "intransitive" verbs. Strictly speaking, the difference between transitive and intransitive verbs has nothing to do with passive. It's a matter of definition, of course. But I think most people would agree that, grammatically speaking, the only difference between transitive and intransitive verbs is that the former takes a noun as a direct object while the latter doesn't. Certainly the reversifier thing and the like some posters explained describe the nature of intransitive verbs well, but it's shouldn't be included in their definition; logical consequences derived from axioms are not axioms.

You might have learned that the direct passive in Japanese was the equivalent of the English passive. This is wrong. Totally wrong. Completely wrong.

The basic principle of the Japanese direct passive is that the object the action of a verb is done on (or simply the subject of a sentence if you will) should be an animated thing that has emotions in a sense. It seems "patient" wildweathel speaks of is referring to this "object of the action" when he talks about direct passive sentences. Typical objects are human beings. In the strict sentence, anything other than humans can't be objects or patients in terms of wildweathel's explanation unless the speaker personifies them to an extent. Your pets are also good examples of possible objects in Japanese direct passive sentences. As this basic principle suggests, the direct passive grammar adds certain emotional nuances to sentences.

In English you can use an emotionless thing as the patient (or the subject) in a passive sentence. And in general, English passive sentences have indirect senses as if a third person is observing situations. For example, the passive form of "I made a mistake" is

A mistake was made.

Here non-animated thing "mistake" is the object of the verb "make," which is very rare in Japanese, and the speaker is less emotionally involved, which is the opposite of the nuance the Japanese direct passive version would have if it were grammatical.

So, if you translate a Japanese passive sentence using English passive, you could grossly mistranslate its nuance. If anything, the regular version of a sentence can be closer to the Japanese direct passive.

As is pointed out by other posters, the "emotion" or "affection" added to a passive sentence can be positive in rare cases. The canonical example of positive cases is 先生にほめられる. The passive version of verb ほめる can add either a positive or negative sense. If anything, positive senses are much more common in this case. You can find long, complicated explanations for this phenomenon if you search the internet in Japanese. But it's not the main point of this thread, so I omit the detail. I guess it might be better to understand that it's the positive nature of the meaning of ほめる.

So the point is:

Japanese direct passive takes animated objects and usually has negative connotations to a degree. Unlike English, passive voice is much rarer than the regular voice. Usually it's more emotional than regular sentences in general. Also, in many cases, the speaker has a reason to choose the rarer grammatical structure.

As is the case with every grammar rule, there are a plethora of exceptions to the basic rules. Generally speaking, the exceptions were introduced by translation of foreign languages. To translate passive sentences that have non-animated objects, not a few Japanese translators have been using "ungrammatical" rules, i.e., they simply borrowed the non-animated subject structure that has the reversed "who did this to whom" relation.

Some expressions got popular to the extent that they are accepted as idiomatic Japanese. Some sound like literal translation that doesn't make much sense. Because they're irregulars introduced from foreign languages, I don't think it's a good idea to try to make sense of them using traditional Japanese grammar. The influx of translation is a very recent phenomenon in the first place.

For some reason many example sentences of direct passive in this thread fall into this irregular category... Well, if your textbook uses this kind of irregular passive voice because it's easier to explain by translation, you might want to burn it and punch the author in the face.

The canonical example of accepted non-animated patient sentences is この会社は2000年に設立された。I don't think this sentence is ungrammatical, and I'm pretty sure anyone who speaks contemporary Japanese would agree. This sentence totally ignores the principle, so the usual affection rule etc. won't work. In fact, this sentence is neutral in spite of the fact that it's in passive voice structure-wise.

Translation-ish grammar is becoming popular, and some native speakers use it in non-translation situations more often than other native speakers. I kind of doubt your average Japanese guy notices this irregular grammar if used in a formal situation; they'd think it's an ambiguous, stiff sentence. I have a lot to say when it comes to this translation grammar, but I think I should move on and explain Japanese indirect passive.

I've written too much about direct passive, so I only explain things no one has mentioned in this thread.

As is often explained in textbooks, indirect passive voice can take intransitive verbs. But not all intransitive verbs can be used. There are two kinds of intransitive verbs. One is called "unergative verbs" (非能格動詞) and the other is "unaccusative verbs" (非対格動詞). Actually English also has this distinction, e.g., "die" is unaccusative, and "run" is unergative. But comparing this kind of thing between words in your mother tongue and foreign languages to "learn" a language is one of the silliest things you can possibly do. Actually 死ぬ is unergative in Japanese while the so-called equivalent word "die" in English is unaccusative. It's like trying to explain to a Chinese guy what's going on in an American football game by using analogy of a ping-pong match. You can't. Well, I think explaining Japanese "direct" passive through English passive is already as silly as this...

Ah, but I digress. when it comes to intransitive verbs, Japanese indirect passive can only take unergative verbs, which are defined as "intransitive verbs whose subjects in regular non-passive voice sentences actively initiate the actions." It might be helpful to think about English sentences "I run," "He resigned," etc. The verbs are intransitive here because they don't have objects in these examples. And "I" and "He" actively initiate the corresponding actions in normal context. So "run" and "resign" are unergative. In English, "die" isn't an action the person who is about to die has control of, so it's not unergative, i.e., it's unaccusative.

Since only unergative verbs (and transitive verbs, of course) are allowed in Japanese indirect passive voice, you can only use intransitive verbs that work like "run," "talk," "resign," etc. (and transitive verbs). The thing is that just because English translations in your dictionary are unergative doesn't mean the Japanese words also have the same grammatical function. As I already said, in the Japanese language, the action "die" is considered to be something actively initiated by the subject who dies. Hence it can be a verb in an indirect passive sentence. In other words, "die" and "死ぬ" are totally different actions while they refer to the exact same thing.

So, um, every single translation-based learning method sucks in this regard. This unergative vs. unaccusative thing appears other grammar points too. Of course, you can't understand them either if you try to understand a foreign language through translation because it's fundamentally wrong.

Anyway, there are two different kinds of indirect passive in Japanese. Reading this thread, I found both types of indirect passive: 太郎は晴子にかばんを盗まれた and 太郎は晴子にタバコをすわれた are indirect passive sentences, but they're grammatically different in usual context. In the former case, かばん is 太郎's, but in usual context, 晴子 smoked her own タバコ, not 太郎's cigarette. Since the "suffering" sense can have different nuances depending on whether the direct object of a verb belongs to the subject, indirect passive sentences should have different kinds of suffering nuances accordingly.

I don't think learning about passive voice and related grammar points would help learn to speak Japanese, so if you're interested, read serious grammar books written for Japanese linguists and/or hardcore grammar Nazis for yourself. These grammar points are pretty much inexplicable if we rely on translation because not only does English have no equivalent grammar, but also one English sentence structure covers different types of passive/passive-like Japanese structure.

Hmm. I think I should stop talking... So, what I wanted to say is, er, what you have just read is pretty much useless because your goal is being fluent in Japanese. Your average Japanese teacher doesn't know this kind of stuff well either. Well, I guess I should have said this earlier, but I doubt anyone would read this post seriously. If you did, you better stop learning about Japanese. If you love grammar, be fluent in Japanese first. That way you can learn grammar by reading academic publications written in Japanese. You can't understand grammar without native fluency anyway.

Edit: Fixed some English grammar. But I don't think I have enough time to read this again and fix all typos and errors...

Last edited by magamo (2009 December 24, 5:24 am)

pm215 wrote:

(When you say this more neutral passive is 'recent' do you mean "since the Meiji era" or "last 20 years" ?)

This started around the Meiji era and is still continuing. "Ungrammatical" translations are being read each day, and some expressions catch on. Elderly Japanese people use more "proper" Japanese than younger generations. My Japanese is more translation-ish than my father's. My supervisor at grad school corrected a sentence in my report in Japanese because he though I copied it from an English book, but I didn't think the sentence was particularly translation-ish. If you talk to a geeky guy who is pretty good at computers, programs, and whatnot, chances are he speaks more translation-ish Japanese than others because the IT industry is full of translations.

No one knows what the Japanese language will be like in 50 years. Maybe translation beats the indigenous Japanese to the extent that passive voice becomes completely neutral. Maybe translation-ish sentences disappear except for already accepted examples. But looking at the huge number of incompetent translators who abuse foreign languages' grammar, I guess the Japanese language will absorb more irregulars. This isn't necessarily a bad thing though.

On a side note, it seems to take a very long time for a translation to be fully integrated. In the Meiji era, great Japanese novelists who wrote classical masterpieces struggled in translating "I love you" into Japanese. Because "love" in this sense was extremely exotic, it was impossible to say it in Japanese. 二葉亭四迷 translated it as 死んでもいい (literally "(I) can die (for you)"). It's said that 夏目漱石 phrased it as 月が綺麗ですね (literally "It's a beautiful moon, isn't it.") because that's what a guy might say when he expresses his "love" to a girl in a romantic way in the context. Some guys in the early Meiji ear started using a Buddhism term 愛 for "love." The meaning was different from "love" or "amour," but it did get popular.

100 years later, "愛してる" is still translation-ish and an exotic concept. It's already integrated in the Japanese language, but still it's a foreign phrase to a large extent. The Japanese have imported more foreign concepts, notions, ideas, and words. Some people started using 恋愛 to describe a romantic relationship in a Western sense. This is a very recent phenomenon. My grandparents wouldn't understand what it means at the same level as younger generations do. This is why Japanese people can't give a clear answer when an English speaker asks a seemingly simple question: "How do I say 'I love you' in Japanese?" You might not be able to comprehend this, but it's an exotic idea in a place where saying "God loves you" to a guy makes the sentence sound as though Jesus was homosexual.

I don't know if it takes longer to integrate an idea than a grammar rule. But I don't think it's surprising that a radical change in grammar like the passive integration is taking more than 100 years.

pm215 wrote:

I dunno, I think remarks about nuance are useful, because it's exactly the kind of thing you can fail to pick up from your reading and never even notice.

In a stricter sense, native speakers can't fail to catch a nuance because what he speaks is part of the language. If some native speakers use an expression in a different way, it's not "wrong" in a (hardcore) descriptive linguistic sense, though it might be so rare that you could say it's below a certain threshold of correctness. Of course this idea is too extreme. If your usage of a phrase is very different from the rest of the native speakers, people would say your understanding is wrong regardless of whether it's your mother tongue. But basically what you say always represents part of the current state of your mother tongue.

My learning strategy is based on this idea. What I think I need to do is consume a huge amount of native material so that people think that what I say also represents part of the language I speak. So there is no such thing as "fail to pick up" per se. I believe there is no clear line that divides native speakers and non-native speakers. If there is, "native speakers" simply refers to people who started learning the language at a very early age, which I think is irrelevant to fluency if "non-native" speakers spent an equivalent amount of time in the language with the same intensity.

Last edited by magamo (2009 December 24, 7:33 am)

magamo wrote:

Translation-ish grammar is becoming popular, and some native speakers use it in non-translation situations more often than other native speakers.

I had a quick scan through a couple of the books I'm currently reading to see what the general use of the passive was. So (unscientific survey alert!):
あなたも殺人犯になれる！(赤川次郎): very little passive at all, both cases in the chapter I looked at were of the animate-subject variety.
ノルウェイの森 (村上春樹): rather more use of the passive, mostly with clearly non-animate subjects. (I'm guessing less passive than you might end up with in a natural English translation, though.)
日本教育小史 (non fiction which seems to me to be written in a fairly formal style; it certainly puts me to sleep) also has a lot of non-animate-subject passives.

I've heard before that Murakami tends to write in what you might call a more translation-ish style, which I guess passive usage would be a part of.

(There may be no clear line between 'native' and 'non-native' but there is a spectrum and I know which end of it I'm on for Japanese :-))

I don't recall being taught that explicitly (though IIRC there are some hints in Jay Rubin's book along those lines), but I'm at my parents and don't have access to any of my books to check.