"the book he had read" in natlangs

[finlay]

Standard Korean (South)

그가 읽은 책
/keu-ga ilg-eun chaek/
3.SG.M-NOM read-PERF.ADJ book

그가 읽었던 책
/keu-ga ilg-eod-deon chaek/
3.SG.M-NOM read-PERF-REFL.ADJ book

Korean works pretty much the same as Japanese.
-던 adds reflective meaning, makes it feel like it happened long ago.

if you want to just use the romanization, it's probably best if you don't put it between // brackets, which denote phonemes – it's conceivable that someone might come along and not recognise that <eo> and <eu> are single vowels. Just put the romanization beneath it, in a similar way to how we've been doing it with Japanese.

[suelior]

if you want to just use the romanization, it's probably best if you don't put it between // brackets, which denote phonemes – it's conceivable that someone might come along and not recognise that <eo> and <eu> are single vowels. Just put the romanization beneath it, in a similar way to how we've been doing it with Japanese.

Thanks for the advise. I put the Yale-romanization beneath it now too.

[merijn]

Standard Korean (South)

그가 읽은 책
ku-ka ilk-un chayk
(keu-ga ilg-eun chaek)
3.SG.M-NOM read-PERF.ADJ book

그가 읽었던 책
ku-ka ilg-ess-ten chayk
(keu-ga ilg-eod-deon chaek)
3.SG.M-NOM read-PERF-REFL.ADJ book

Korean works pretty much the same as Japanese.
-던 adds reflective meaning, makes it feel like it happened long ago.

Mandarin
他讀過的書
tā dúguò de shū
3s.M read-already GEN book

過 indicates that the action has already been performed.
他讀過 'He already read' is treated as a modifier for the noun 書 'book', so the genitive 的 is used.

I always assumed that you can leave out the subject in Chinese and Korean, like you can do in Japanese, so that the sentences could be:

dúguò de shū (Mandarin)

and

ilk-un chayk (Korean) or
ilg-ess-ten chayk (Korean)

Is that correct? Or is that not a natural translation of my phrase?

[Astraios]

I always assumed that you can leave out the subject in Chinese and Korean, like you can do in Japanese, so that the sentences could be:

dúguò de shū (Mandarin)

Not in Chinese.

[merijn]

I always assumed that you can leave out the subject in Chinese and Korean, like you can do in Japanese, so that the sentences could be:

dúguò de shū (Mandarin)

Not in Chinese.

I did a search and found out that Chinese is a pro-drop language. Wikipedia (but it is not just Wikipedia who makes the claim) gives the following sentences:
bù zhī dào. xĭ huān ma?
Not know. like [QUESTION MARKER]?
"I don't know. Do you like it?"

Is there something in relative clauses that blocks the pro-drop?

[Astraios]

I did a search and found out that Chinese is a pro-drop language. Wikipedia (but it is not just Wikipedia who makes the claim) gives the following sentences:
bù zhī dào. xĭ huān ma?
Not know. like [QUESTION MARKER]?
"I don't know. Do you like it?"

Is there something in relative clauses that blocks the pro-drop?

Dropping the pronoun doesn't let the book be identified - since the verb doesn't show person it'd mean "the book ... had read" - and it doesn't really sound quite like something you can say, because the relative clause's purpose is to identify which book the speaker means.

[merijn]

But the same is true for Japanese, and yet you can leave out the subject. What about the following context: it is in a story with the following sentences "he read it in a book, but the book he had read seemed to have disappeared" can out leave out the subject there?

[Astraios]

I don't think it sounds right. I'm not a native speaker though, so I may be wrong, but it doesn't feel right.

[treskro]

But the same is true for Japanese, and yet you can leave out the subject. What about the following context: it is in a story with the following sentences "he read it in a book, but the book he had read seemed to have disappeared" can out leave out the subject there?

他在一本書裡面讀過，但他讀的那本書好象不見了。
3s.M LOC one COUNTER book within read already, but 3s.M read GEN that COUNTER book seem disappear PERF

This is valid, but dropping the subject is also valid, since the 那 "that" directly specifies which book is being referred to, namely, "the book in which he read it". This leaves us with
他在一本書裡面讀過，但那本書好象不見了。
3s.M LOC one COUNTER book within read already, but that COUNTER book seem disappear PERF

他讀過的書 ta duguo de shu and 讀過的書 duguo de shu imply two different things. The first one specifies book(s) that he has read, while the second refers more generally to books that any single person has read.

[suelior]

Standard Korean (South)

그가 읽은 책
ku-ka ilk-un chayk
(keu-ga ilg-eun chaek)
3.SG.M-NOM read-PERF.ADJ book

그가 읽었던 책
ku-ka ilg-ess-ten chayk
(keu-ga ilg-eod-deon chaek)
3.SG.M-NOM read-PERF-REFL.ADJ book

Korean works pretty much the same as Japanese.
-던 adds reflective meaning, makes it feel like it happened long ago.

Mandarin
他讀過的書
tā dúguò de shū
3s.M read-already GEN book

過 indicates that the action has already been performed.
他讀過 'He already read' is treated as a modifier for the noun 書 'book', so the genitive 的 is used.

I always assumed that you can leave out the subject in Chinese and Korean, like you can do in Japanese, so that the sentences could be:

dúguò de shū (Mandarin)

and

ilk-un chayk (Korean) or
ilg-ess-ten chayk (Korean)

Is that correct? Or is that not a natural translation of my phrase?

In Korean you can. In the library we sometimes have the book-returning box(I don't know what it's called) with a label saying "다 읽은 책은 여기에."

다 읽은 책은 여기에
ta ilg-un chayg-un ye-gi-ey
(ta ilg-eun chaeg-eun yeo-gi-e)
All(adv) read-PERF.ADJ book-LIM.PTCL here-LOC

In this case, it could be a book (or books) read by anyone or any number of people. It doesn't matter who read it, and presumably it is you who read it (unless you're returning books your friend borrowed) so it can be omitted.
In Korean you can drop pretty much anything if the context is enough to tell the listener what you are reffering to.

[Magb]

Standard Swedish:

Boken, [som] han [hade] läst.
book-DEF [that] he [had] read-SUPINE

Same structure for all of Northern Germanic.

Boka, som han hadde lest.
Bogen, der han havde læst.
Bókin, sem hann hafði lesið.
Bókin, sum hann hevði lisið.

You just can't let go of your precious commas, can you? Keep in mind that "Boken, som han hade läst" has a subtly different meaning than "Boken som han hade läst", and it's the latter that's the correct translation of the OP's phrase.

It probably also bears mentioning that while all of these languages can drop the relative pronoun, only Swedish can drop the auxiliary "had".

[Izambri]

Catalan

El llibre que ell ha llegit
The book that he has read.

[finlay]

Standard Swedish:

Boken, [som] han [hade] läst.
book-DEF [that] he [had] read-SUPINE

Same structure for all of Northern Germanic.

Boka, som han hadde lest.
Bogen, der han havde læst.
Bókin, sem hann hafði lesið.
Bókin, sum hann hevði lisið.

You just can't let go of your precious commas, can you? Keep in mind that "Boken, som han hade läst" has a subtly different meaning than "Boken som han hade läst", and it's the latter that's the correct translation of the OP's phrase.

Hey Sko, why don't you actually look up the conventional usage of commas in each language that you translate into? Including your own....

[Magb]

Hey Sko, why don't you actually look up the conventional usage of commas in each language that you translate into? Including your own....

Maybe I shouldn't be too presumptuous about the use of the comma in Swedish, but in Norwegian the presence of the comma pretty clearly indicates the reading "the book, which he had read", as opposed to "the book (that) he had read". Not a major difference, but not a completely negligible one either.

[Whimemsz]

Ojibwe would be something like iw mazina'igan gaa-agindamaan, where iw (with many variants) is the proximal inanimate demonstrative, mazina'igan is "book", and gaa-agindamaan contains the verb -agindam- "read something inanimate", the past tense proclitic gii-, and the suffix -aan which marks first person singular acting on an inanimate object in the conjunct order (basically, dependent clauses). The whole verbal complex is relativized by applying an ablaut process termed "initial change", which in this case converts the /i:/ in the past tense prefix to /a:/; initial change has several different functions, but one is forming relative constructions.

[Astraios]

Maybe I shouldn't be too presumptuous about the use of the comma in Swedish, but in Norwegian the presence of the comma pretty clearly indicates the reading "the book, which he had read", as opposed to "the book (that) he had read". Not a major difference, but not a completely negligible one either.

My syntax lecturer (and me) would disagree with that it's not a major difference. "The students, who cheated, will be punished" vs. "The students who cheated will be punished" is a pretty important distinction.

[Skomakar'n]

Standard Swedish:

Boken, [som] han [hade] läst.
book-DEF [that] he [had] read-SUPINE

Same structure for all of Northern Germanic.

Boka, som han hadde lest.
Bogen, der han havde læst.
Bókin, sem hann hafði lesið.
Bókin, sum hann hevði lisið.

You just can't let go of your precious commas, can you? Keep in mind that "Boken, som han hade läst" has a subtly different meaning than "Boken som han hade läst", and it's the latter that's the correct translation of the OP's phrase.

Hey Sko, why don't you actually look up the conventional usage of commas in each language that you translate into? Including your own....

Because I'm not placing them according to convention. I'm placing them according to personal preference. e;
Danish (and possibly Faroese) is probably the only one where this would be really conventional in this case, though.

Hey Sko, why don't you actually look up the conventional usage of commas in each language that you translate into? Including your own....

Maybe I shouldn't be too presumptuous about the use of the comma in Swedish, but in Norwegian the presence of the comma pretty clearly indicates the reading "the book, which he had read", as opposed to "the book (that) he had read". Not a major difference, but not a completely negligible one either.

Yeah. I really need a context to know for sure. Maybe the OP actually specified one. Not sure. I think it adds a nice division when showing an example like this, at least.

[Qwynegold]

I'm crap at making interlinears for natlangs, but something like this for Finnish:

kirja jonka hän oli lukenut
kirja jonka hän ol-i luke-nut
book REL.ACC 3SG be-PST.3SG read-PST.ACT.PCP

That's a past active participle suffix on read.

[finlay]

Because I'm not placing them according to convention. I'm placing them according to personal preference. e;
Danish (and possibly Faroese) is probably the only one where this would be really conventional in this case, though.

Well stop it. Commas mean things.

[Skomakar'n]

Because I'm not placing them according to convention. I'm placing them according to personal preference. e;
Danish (and possibly Faroese) is probably the only one where this would be really conventional in this case, though.

Well stop it. Commas mean things.

Which is exactly why I use them. The lack of them takes away meaning and clarity!

[Astraios]

Which is exactly why I use them. The lack of them takes away meaning and clarity!

No, using them where they shouldn't be takes it away. If you always write "the students, who cheated, will be punished", how does anyone know when you mean "the students who cheated will be punished"? They're different things!

[ná'oolkiłí]

Georgian:

წიგნი, რომელიც (მან) წაიკითხა
c̣ign-i romel-i-c (man) c̣a-i-ḳitx-a
book-NOM which-NOM-REL (3S.NAR) PV-VV-read-AOR.3S

PV is "preverb" and VV is "version vowel". The first generally encodes perfective aspect, but, for this verb, the second morpheme doesn't really have a function that is easy to pin down—it basically is just part of the aorist conjugation. The relative pronoun reflects the case of embedded pivot (t in "the book that he read t"), here nominative because read in the aorist calls for a nominative direct object. The pronoun is optional. NAR is "narrative", the traditional Kartvelological name for the case often (and somewhat misleadingly, IMO) called "ergative".

Georgian has a second, more interesting way of forming relative clauses, but this strategy doesn't really work as a bare DP, so I'll put it in a sentence.

წიგნი რომ წაიკითხეს, ლევანმა ის იყიდა.
c̣igni rom c̣aiḳitx-es, Levan-ma is v-i-qid-e
book.NOM REL read-3P.AOR, Levan-NAR 3S.NOM VV-buy-3S.AOR
"Levan bought the book they wrote"

This type of relative clause is internally-headed and separated from the matrix pivot. The invariant complementizer rom is usually second (it might actually be a clitic that attaches to the embedded pivot, but I haven't looked into it). This whole relative clause is placed at the sentence periphery, and a resumptive pronoun stands where the matrix pivot usually would. I changed the subject of "write" to avoid potential ambiguity.

I think it's most natural to translate English pluperfect into Georgian as the aorist (perfective past), but I suppose you could use either the perfect or pluperfect. However, these don't really have the function their names suggest, and they carry certain nuances. They also govern inversion, which means the logical subject is marked as the morphological indirect object, and the direct object as the subject. In the perfect you'd have the following forms, which convey evidentiality:

წიგნი, რომელიც (მას) წაუკითხია
c̣igni, romelic (mas) c̣a-u-ḳitx-ia
book.NOM, which.NOM.REL (3S.DAT) PV-3S.IO-read-3.PERF
"The book that he [apparently] wrote"

წიგნი რომ წაუკითხიათ, ლევანმა ის იყიდა.
c̣igni rom c̣a-u-kitx-ia-<t>, Levanma is iqida
— — PV-3<P>.IO-read-3.PERF — — —
"Levan bought the book they [apparently] wrote"

As for the pluperfect, on its own, it conveys unexpectedness. (NB: I'm not entirely sure I conjugated the verb right, and I think there might be syntactic constraints on the pluperfect's usage... the perfect tenses are pretty advanced material that I haven't really learned yet)

წიგნი, რომელიც (მას) წაეკითხა
c̣igni, romelic (mas) c̣a-Ø-e-ḳitx-a
book, which (3S) PV-3S.IO-VV-read-3.PLU
"The book that he wrote/had written/managed to write after all"

წიგნი რომ წაეკითხათ, ლევანმა ის იყიდა.
c̣igni rom c̣a-Ø-e-kitx-a-<t>, Levanma is iqida
— — PV-3<P>.IO-VV-read-3.PLU — — —
"Levan bought the book they wrote/had written/managed to write after all"

[Skomakar'n]

Which is exactly why I use them. The lack of them takes away meaning and clarity!

No, using them where they shouldn't be takes it away. If you always write "the students, who cheated, will be punished", how does anyone know when you mean "the students who cheated will be punished"? They're different things!

I don't always write it like that. ;3
Like I said, I just felt it would be clearer for this small example, since I didn't know what the context was.

I'd probably write 'boken som han hade läst var spännande' ('the book that he had read was exciting') a lot of the time, but I might as well write 'boken, som han hade läst, var spännande'. It all depends on the context and my mindset and mood when I'm writing. I don't think it makes more than a stylistic difference in the specific case exemplified here, while it can indeed actually change the meaning in some other cases. In this one, not at all.

[finlay]

Which is exactly why I use them. The lack of them takes away meaning and clarity!

No, using them where they shouldn't be takes it away. If you always write "the students, who cheated, will be punished", how does anyone know when you mean "the students who cheated will be punished"? They're different things!

I don't always write it like that. ;3
Like I said, I just felt it would be clearer for this small example, since I didn't know what the context was.

I'd probably write 'boken som han hade läst var spännande' ('the book that he had read was exciting') a lot of the time, but I might as well write 'boken, som han hade läst, var spännande'. It all depends on the context and my mindset and mood when I'm writing. I don't think it makes more than a stylistic difference in the specific case exemplified here, while it can indeed actually change the meaning in some other cases. In this one, not at all.

Look, ok, technically I don't know about Swedish (although I get the impression from what others are saying that it exemplifies a similar distinction in the written standard), but in English, the sentences
the book which he'd read was exciting
and
the book, which he'd read, was exciting
mean different things. Only in the first type can you replace 'which' with 'that' or a zero-complementizer. In the second sentence, we already know which book the writer is referring to from the wider context. We're talking about one specific book here. We then add the extra information – the fact that a certain man had read this book – as a non-restrictive relative clause separated by commas. We don't know about the existence of any other books, or whether the man has read them.

In the first sentence, we already know from the wider context about a set of books. We then use a restrictive relative clause, not separated by commas, to indicate that we are choosing one book out of this set, and that the man referred to by 'he' has read this specific book and no others. We use the fact that he's read the book to distinguish it from other books.

They're. Different.

Because of the grammatical structure of the OP's English sentence (doesn't use commas and has a zero-complementizer), we know that it must be type 1, the restrictive relative clause.

http://en.wikipedia.org/wiki/Restrictiveness

[Skomakar'n]

How would the difference be expressed in speech, if all words are the same?