Dear Francesc González i Planas

1)

"1) Following Sato (2012), embedded questions in (i) and (ii) can be interpreted as indirect discourses.

(i) Taroo-ga Ziroo-ni [Hanako-ga nani-o kat-ta ka to] tazune-ta

(ii) Taroo-ga Ziroo-ni [Hanako-ga kare-no hon-o kat-ta ka to] tazune-ta."

First, I want to confirm  the term "indirect discourses". Is the term "indirect discourse" the same as "indirect speech”? If so, I don’t think that the interpretation of Sato (2012) is correct. The problem is the existence of a particle “to”, which functions as a marker of citation as Professor Hyo Lee pointed out in the response to my answer to your question. In fact, in the examples of some articles about the Japanese indirect question that I read, the particle “to” is not used but the particle “o” , marker of a direct object (accusative marker). According to the articles in question, the examples (i)and (ii) will be:

(i’) Taroo-ga Ziroo-ni [Hanako-ga nani-o kat-ta ka] o tazune-ta.

(ii’) Taroo-ga Ziroo-ni [Hanako-ga kare-no hon-o kat-ta ka] o tazune-ta.

I have another doubt about the interpretation of Sato (2012). Following Sato (2012),. the phrase [kare-no hon] in (ii) refers to “Taro’s book”. But according to my native intuition, [kare] can refer not only to Taro but also to Ziroo and someone else.  

One more thing. To me the examples (i) and (ii) sound a little strange because in a natural Japanese the human subject in the principle clause is not followed, in general, by the particle “ga” but by the particle “wa” that functions as a theme marker. That is, the examples (i) and (ii) will sound more natural if they are changed to;

(i”) Taroo-wa Ziroo-ni [Hanako-ga nani-o kat-ta ka] o tazune-ta.

(ii”) Taroo-wa Ziroo-ni [Hanako-ga kare-no hon-o kat-ta ka] o tazune-ta.

2)

(iii) [Kare-no hon-o] Taroo-ga Ziroo-ni [dare-ga ___ kat-ta ka to] tazune-ta. ‘Taro asked Ziroo who bought his book.’

(iv) Taroo-wa [kare-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka to] tazune-ta.  ‘Taro asked Ziroo if Hanako bought his book

First, I don’t think that the examples (iii) and (iv) are indirect questions because there is a particle “to” in the subordinate clauses “dare-ga ___ kat-ta ka to” and “Hanako-ga ___ kat-ta ka to”.  The correct indirect questions will be;

(iii’) [Kare-no hon-o] Taroo-wa(ga) Ziroo-ni [dare-ga ___ kat-ta ka] o tazune-ta.

(iv’) Taroo-wa [kare-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka] o tazune-ta.  ‘Taro asked Ziroo if Hanako bought his book

Interestingly, the examples (iii’) and (iv’) don’t sound well (at least, I will take some time to understand them correctly), if not ungrammatical.

However, the next examples (iii”) and (iv”) which are interpretated as direct questions don’t sound so bad.

(iii”) [Kare-no hon-o] Taroo-wa(ga) Ziroo-ni [dare-ga ___ kat-ta ka] to tazune-ta. ‘Taro asked Ziroo who bought his book.’

(iv’) Taroo-wa [kare-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka] to tazune-ta.  ‘Taro asked Ziroo if Hanako bought his book

I hope my response will be useful for you.

Hiromi Yamamura

 Scrambling as in these sentences is discussed in Saito (1992), Journal of East Asian Linguistics, 1, 69-118.

Dear Francesc González i Planas

1)

"1) Following Sato (2012), embedded questions in (i) and (ii) can be interpreted as indirect discourses.

(i) Taroo-ga Ziroo-ni [Hanako-ga nani-o kat-ta ka to] tazune-ta

(ii) Taroo-ga Ziroo-ni [Hanako-ga kare-no hon-o kat-ta ka to] tazune-ta."

First, I want to confirm  the term "indirect discourses". Is the term "indirect discourse" the same as "indirect speech”? If so, I don’t think that the interpretation of Sato (2012) is correct. The problem is the existence of a particle “to”, which functions as a marker of citation as Professor Hyo Lee pointed out in the response to my answer to your question. In fact, in the examples of some articles about the Japanese indirect question that I read, the particle “to” is not used but the particle “o” , marker of a direct object (accusative marker). According to the articles in question, the examples (i)and (ii) will be:

(i’) Taroo-ga Ziroo-ni [Hanako-ga nani-o kat-ta ka] o tazune-ta.

(ii’) Taroo-ga Ziroo-ni [Hanako-ga kare-no hon-o kat-ta ka] o tazune-ta.

I have another doubt about the interpretation of Sato (2012). Following Sato (2012),. the phrase [kare-no hon] in (ii) refers to “Taro’s book”. But according to my native intuition, [kare] can refer not only to Taro but also to Ziroo and someone else.  

One more thing. To me the examples (i) and (ii) sound a little strange because in a natural Japanese the human subject in the principle clause is not followed, in general, by the particle “ga” but by the particle “wa” that functions as a theme marker. That is, the examples (i) and (ii) will sound more natural if they are changed to;

(i”) Taroo-wa Ziroo-ni [Hanako-ga nani-o kat-ta ka] o tazune-ta.

(ii”) Taroo-wa Ziroo-ni [Hanako-ga kare-no hon-o kat-ta ka] o tazune-ta.

2)

(iii) [Kare-no hon-o] Taroo-ga Ziroo-ni [dare-ga ___ kat-ta ka to] tazune-ta. ‘Taro asked Ziroo who bought his book.’

(iv) Taroo-wa [kare-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka to] tazune-ta.  ‘Taro asked Ziroo if Hanako bought his book

First, I don’t think that the examples (iii) and (iv) are indirect questions because there is a particle “to” in the subordinate clauses “dare-ga ___ kat-ta ka to” and “Hanako-ga ___ kat-ta ka to”.  The correct indirect questions will be;

(iii’) [Kare-no hon-o] Taroo-wa(ga) Ziroo-ni [dare-ga ___ kat-ta ka] o tazune-ta.

(iv’) Taroo-wa [kare-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka] o tazune-ta.  ‘Taro asked Ziroo if Hanako bought his book

Interestingly, the examples (iii’) and (iv’) don’t sound well (at least, I will take some time to understand them correctly), if not ungrammatical.

However, the next examples (iii”) and (iv”) which are interpretated as direct questions don’t sound so bad.

(iii”) [Kare-no hon-o] Taroo-wa(ga) Ziroo-ni [dare-ga ___ kat-ta ka] to tazune-ta. ‘Taro asked Ziroo who bought his book.’

(iv’) Taroo-wa [kare-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka] to tazune-ta.  ‘Taro asked Ziroo if Hanako bought his book

I hope my response will be useful for you.

Hiromi Yamamura

 Dear Hiromi Yamamura,

Thank you very much. Your response is very useful for me, but I have some clarifications and comments.

Clarifications:

The examples in (i)-(iv) are created by me following Saito's ideas that the complementizer to can select indirect speech, because his original examples are not useful for my purposes. According to my little knowledge on Japanese, I'm agreeing with you that sentences headed by a topic are more natural than

headed ones by a nominative subject. (I have read that, if there is not a topic, it is needed add the complementizer koto.) Another mistake in (ii) is "his book = Taro's book", like it is pointed by you. The correct sentences, I think, are the following:

(1)

Taroo-wa Ziroo-ni [Hanako-ga nani-o kat-ta ka to] tazune-ta.

Taro-TOP Ziroo-DAT [Hanako-NOM what-ACC buy-PST Q C] inquire-PST

‘Taro asked Ziroo what Hanako bought.’

(2)

Taroo-wa [zibun-no hon-o Hanako-ga kat-ta ka to] Ziroo-ni tazune-ta.

Taro-TOP [self-GEN book-ACC Hanako-NOM buy-PST Q C] Ziroo-DAT inquire-PST

‘Taro asked Ziroo if Hanako bought his book.’ [his book = Taro's book]

Comments:

Saito's (2012) paper compares embedded interrogative and imperative clauses headed by to with Spanish embedded interrogatives and commands headed by que (Plann 1982, Rivero 1994). The following examples are Spanish versions of my (1) and (2), if my interpretation of Saito's paper is correct:

(3)

Juan le {dijo/preguntó} a Pedro que qué compró María.

Juan CL.3.DAT {said/asked} to Pedro that what bought María

`Juan asked Pedro what María bought.' ~ `Juan asked Pedro: "What did María buy?"'

(4)

Juani le {dijo/preguntó} a Pedroj que si María compró sui/j/k libro.

Juan CL.3.DAT {said/asked} to Pedro that if María bought his book

`Juan asked Pedro if María bought {his/her/their} book.' ~ `Juan asked Pedro: "Did María buy {my/your/his/her/their} book?"'

The embedded clauses of (3) and (4) are opaque domains for long-distance movement (focus, theme, or wh-

phrases extraction):

(4')

* Su libro, Juan le dijo/preguntó a Pedro que si María lo compró.

his/her/their book Juan CL.3.DAT said/asked to Pedro that if María CL.3.ACC bought

Intended: `Juan asked Pedro if María bought his book.'

In (4') movement is not possible because it is a sentence with an embedded interrogative in indirect speech, but in (4'') ---a sentence with an (embedded) indirect question--- the movement is available:

(4'')

Su libro, Juan le preguntó a Pedro si María lo compró.

his/her/their book Juan CL.3.DAT said/asked to Pedro that if María CL.3.ACC bought

`Juan asked Pedro if María bought his book.'

My hipothesis is that examples like (1) or (2) can be interpreted as sentences with an embedded interrogative in indirect speech like (3) and (4) in Spanish. If the indirect speech interpretation is good for (1) and (2), and they have the same properties that Spanish clauses, then long-distance scrambling should not be possible in sentence like (iii) and (iv), here (5) and (6).

(5)

Taroo-wa [zibun-no hon-o] Ziroo-ni [dare-ga ___ kat-ta ka to] tazune-ta.

‘Taroi asked Ziro who bought hisi book.’

(6)

Taroo-wa [zibun-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka to] tazune-ta.

‘Taroi asked Ziro if Hanako bought hisi book.’

If you wrote "examples (iii”) and (iv”) which are interpretated as direct questions don’t sound so bad", do you consider examples (5) and (6) as grammatical if they are interpretated as indirect speech? Is it possible the indirect speech interpretation in sentences like (1) or (2)?

Thank you very much.

Francesc

REFERENCES

Plann, Susan (1982). Indirect questions in Spanish. LI 13(2):297-312.

Rivero, María-Luisa (1994). On indirect questions, commands, and Spanish quotative 'que'. LI 25(3):547-554.

Saito, Mamoru (2012). Sentence types and the Japanese right periphery. In: Discourse and grammar: From sentence types to lexical categories, edited by Günter Grewendorf & Thomas Ede Zimmermann, 147-175. Berlin: De Gruyter.

Dear Franscesc:

    Thank you for your reply. I don’t know very well Japanese generative grammar, but following some articles that I could read, there are two points of view about the stand of the particle “to”. One is a point of view, according to which the particle “to” is a complementizer, whereas the other is the one, folloing which the same particle “to” is not a complementizer. Since it is almost impossible for me to decide which point of view is adquate, I have decided to describe a real and concrete behavior of the particle “to” that is found both in direct speech and in indirect speech. See the examples below:

 1. Taro  le    dijo  a          Jiro: “Hanako     vive       en          Tokio.”

Taro-wa kare-ni it-ta                 Hanako-wa  sunde-iru  ni(particle)

     The example 1 is a Spanish assertive sentence in direct speech. Here is not a Spanish complementizer “que”. However, in the corresponding Japanese sentences appears the particle “to” as the translations (a) and (b) show. All the Japanese sentences shown below are the ones in direct speech because it is possible to insert -mas- form in the embedded assertive sentence.

lit.: a. Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-iru to it-ta.

 a’. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-iru φ it-ta.

a”. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-iru o it-ta.

   b. Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-i mas-u to it-ta.

b’. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-i mas-u  φ it-ta.

b”. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-i mas-u o it-ta.

2. Taro dijo a Jiro que Hanako vive(vivía) en Tokio

         The example 2 is a Spanish assertive sentence in indirect speech. Here the Spanish complementizer “que” is indispensable. The same thing occurrs also in the corresponding Japanese sentences as the examples below show. In Japanese indirect speech it is impossible to insert -mas- form in the embedded assertive sentence because the –mas- form is a marker of the speaker’s actittude.

lit.: a. Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-iru to it-ta.

   a’. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-iru φ it-ta.

   a”. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-iru o it-ta.

   b. *Taro-wa Jiro-ni Hanako-wa Tokio-ni sunde-i mas-u to it-ta. (in the case of the interpretation of indirect speech)

         Now I want you to pay attention to the examples 1(a) and 2(a) because both of them need the particle “to” for being grammatical. That is, the presence or not of the particle “to” is not clue to distinguish a direct speech from an indirect speech.

3. Taro    le      dijo : “¿Hanako    vive          en Tokio ?”

Taro-wa kare-ni  it-ta  Hanako-wa  sunde-iru     ni(particle)

        The example 2 is a Spanish interrogative sentence in direct speech. Here is not a Spanish complementizer “que” neither. However, in the corresponding Japanese sentences appears the particle “to” as its translations (a) and (b) show.

lit.: a. Taro-wa kare-ni Hanko-wa Tokio-ni sunde-iru [ka] to it-ta

   a’. *Taro-wa kare-ni Hanko-wa Tokio-ni sunde-iru [ka] φ it-ta.

   a”. *Taro-wa kare-ni Hanko-wa Tokio-ni sunde-iru [ka] o it-ta.

   b. Taro-wa kare-ni Hanko-wa Tokio-ni sunde-i mas-u [ka] to it-ta.

4. *Taro le     preguntó      que　Hanako     vive       en Tokio.

       kare-ni kii-ta/tazune-ta       hanako-wa  sunde-iru  ni(particle)

lit.: a. Taro-wa kare-ni Hanako-wa Tokio-ni sunde-iru [ka] to kii-ta 135/tazune-ta 122

  a’. Taro-wa kare-ni Hanako-wa Tokio-ni sunde-iru [ka]  φ kii-ta 245/tazune-ta. 170

  a”. Taro-wa kare-ni Hanako-wa Tokio-ni sunde-iru [ka] o kii-ta 241 / tazune-ta 119

　　  The Spanish interrogative sentence in indirect speech like the example 4, in which the complementizer “que” is inserted, is not permitted.  The correct Spanish interrogatice sentence in indirect speech always needs a complementizer “si” as the later example 5 shows. By contrast, in Japanese there is no difference between the interrogative sentence in direct speech like the example 3(a) and the interrogative sentence in indirect speech like the example 4(a), apart from their main verbs. Also it should be paid attention to the fact that, when the verbs “kiku”/ “tazuneru” that are equivalent to the Spanish verb “preguntar” are used as main verbs, the particle “to” can not only be omitted as the example 4(a’) shows but also can be substitued for by the particle “o” that functions as a marker of direct object as the example 4(a”) shows. And what we cannot overlook here is that, according to the result of the search in Google, the possiblity of the appearance of the particle “to” is less than that of the omission of the particle “to” and that of the appearance of the particle “o” as the numbers of the examples 4(a), 4(a’) and 4 (a”) show. It shows that the possibility of the appearance of the particle “to” depends on the meaning of the preceding verbs because, when the preceding verb is reporting verb like “iu” (“decir =say”), the particle “to” is indispensable whichever the speech type in which the sentence in question is expressed is, whereas when the preceding verb is a kind of interrogation like “kiku/tazuneru” (“preguntar=ask”), the particle “to” tends to be omitted.

5. Taro le     preguntó [que] si Hanako vive(vivía) en Tokio.

                         kadouka

a. Taro-wa kare-ni Hanako-ga Tokio-ni sunde-iru [ka] douka [to] kii-ta. 27/tazune-ta    21

a’. Taro-wa kare-ni Hanako-ga Tokio-ni sunde-iru [ka] doukaφkii-ta. 208/tazune-        ta253

a”. Taro-wa kare-ni Hanako-ga Tokio-ni sunde-iru [ka] douka o kii-ta. 166/tazune-ta 137

       The example 5 is a Spanish sentence with an embedded interrogative in indirect speech, which can be tranlated to the Japanese examples 5(a), 5(a’) and 5(a”). But, according to the result of the serach in Google, the particle “to” doesn’t appear so frequently as its omission (=φ) and the particle “o” .

7. Taro   le       dijo : “¿Qué   compró Hanako ?”

Taro-wa  kare-ni  it-ta   nani-o  kat-ta  Hanako-wa

lit.: a. Taro-wa kare-ni Hanako-wa nani-o kat-ta [ka] to it-ta

   a’. *Taro-wa kare-ni Hanako-wa nani-o kat-ta [ka] φ it-ta.

   a”. *Taro-wa kare-ni Hanako-wa nani-o kat-ta [ka] o it-ta.

   b. Taro-wa kare-ni Hanako-wa nani-o kai mas-i-ta [ka] to it-ta.

           The example 7 is a Spanish interrogative sentence in direct speech. Here is also not a Spanish complementizer “que”. However, in the corresponding Japanese sentences appears the particle “to” as the examples (a) and (b) show.

8. Taro le       preguntó      [que] qué   compró  Hanako.

Taro-wa kare-ni  kii-ta/tazune-ta      nani-o  kat-ta  Hanako-wa

lit.: a. Taro-wa kare-ni Hanako-wa(ga) nani-o kat-ta [ka] [to] kii-ta /tazune-ta

   a’. Taro-wa kare-ni Hanako-ga(wa) nani-o kat-ta [ka] φ kii-ta /tazune-ta.

   a”. Taro-wa kare-ni Hanako-ga(wa) nani-o kat-ta [ka] o kii-ta. /tazune-ta

        The example 8 is a Spanish sentence with an embedded interrogative in indirect speech. In this sentence the complementizer “que” is not indispensable but can be omitted.  The Spanish example 8 can be translated to the Japanese examples (a), (a’) and (a”). That is, the particle “to” is also not indispensable here. In the examples (a), (a’) and (a”), there is one more interesting phenomenon, that is, the change of the particles “wa” and “ga” in the subject Hanako that appears in the embedded interrogative. At least, when the particle “to” appears after the embedded interrogative, Hanako-wa sounds better than Hanako-ga at least to me. By contrast, when the particle “to” doesn’t appear after the embedded interrogative, Hanako-ga sounds much better than Hanako-wa to me. I think that it suggests that the example (a) is a kind of citation, whereas the examples (a’) and (a”) are not so. (Cf. In the Japanese literature, it is well known that the subject in a subordinate clause which is not a citation takes obligatorily the particle “ga” . )

       I don’t know if what I have pointed out below can be useful for your questions. What I want to say with special emphasis is that the real behavior of the supposed Japanese complementizer “to” is very different from that of its supposed corresponding Spanish complementizer “que”.  It will be because of this difference that it is very difficult to distinguish a sentence in direct speech from a sentence in indirect speech in Japanese.

 Hiromi Yamamura

 P.D.:

(3)

Juan le {dijo/preguntó} a Pedro que qué compró María.

Juan CL.3.DAT {said/asked} to Pedro that what bought María

`Juan asked Pedro what María bought.' ~ `Juan asked Pedro: "What did María buy?"'

lit.: a. Juan-wa Pedro-ni María-ga nani-o kat-ta [ka] to kii-ta 0/tazune-ta 0

   a’. Juan-wa Pedro-ni María-ga nani-o kat-ta [ka] φ kii-ta 8/tazune-ta.2

   a”. Juan-wa Pedro-ni María-ga nani-o kat-ta [ka] o kii-ta. 0/tazune-ta 0

(4)

Juani le {dijo/preguntó} a Pedroj [que] si María compró sui/j/k libro.

Juan CL.3.DAT {said/asked} to Pedro that if María bought his book

`Juan asked Pedro if María bought {his/her/their} book.' ~ `Juan asked Pedro: "Did María buy {my/your/his/her/their} book?"'

a. Juan-wa Pedro-ni María-ga zibun-no hon-o kat-ta [ka] douka [to] kii-ta. 27/tazune-ta 21

a’. Juan-wa Pedro-ni María-ga zibun-no hon-o kat-ta [ka] doukaφkii-ta.208/tazune-ta 253

a”. Juan-wa Pedro-ni María-ga zibun-no hon-o kat-ta [ka] douka o kii-ta. 166/tazune-ta 137

(4')

* Su libro, Juan le dijo/preguntó a Pedro [que] si María lo compró.

his/her/their book Juan CL.3.DAT said/asked to Pedro that if María CL.3.ACC bought

Intended: `Juan asked Pedro if María bought his book.'

a. i. Juan-wa Pedro-ni María-ga(wa) zibun-no hon-o kat-ta [ka] douka [to] kii-ta /tazune -ta  María-ga: zibun-no hon=(1)Juan’s book,(2) María’s book,  María-wa: zibun-no hon=(1)María’s book, (2)Juan’s book

ii. zibun-no hon-o Juan-wa Pedro-ni María-ga(wa) kat-ta [ka] douka [to] kii-ta /tazune –ta  María-ga: zibun-no hon=Juan’s book, María-wa:zibun-no hon=Juan’s book

a’. i. Juan-wa Pedro-ni María-ga zibun-no hon-o kat-ta [ka] doukaφkii-ta./tazune-ta

     zibun-no hon=Juan’s book

  ii. zibun-no hon-o Juan-wa Pedro-ni María-ga  kat-ta [ka] doukaφkii-ta./tazune-ta

zibun-no hon=Juan’s book

a”. i. Juan-wa Pedro-ni María-ga zibun-no hon-o kat-ta [ka] douka o kii-ta. /tazune-ta

   zibun-no hon=Juan’s book

ii.  zibun-no hon-o Juan-wa Pedro-ni María-ga kat-ta [ka] douka o kii-ta. /tazune-ta

   zibun-no hon=Juan’s book

 (5)

Taroo-wa [zibun-no hon-o] Ziroo-ni [dare-ga ___ kat-ta ka to] tazune-ta.

‘Taroi asked Ziro who bought hisi book.’

      This sentence is gramamtical.

(6)

Taroo-wa [zibun-no hon-o] Ziroo-ni [Hanako-ga ___ kat-ta ka to] tazune-ta.

‘Taroi asked Ziro if Hanako bought hisi book.’

      This sentence is also grammatical.

Dear Hiromi Yamamura,

Thanks for your answer. There are very interesting data that I should consider carefully. This kind of sentences in Japanese and Spanish have interesting points of contact that need to be analyzed cross-linguistically.