尊敬表現における名詞化について

全文

(1)Title. 尊敬表現における名詞化について. Author(s). 大賀, 京子. Citation. 北海道教育大学紀要. 人文科学・社会科学編, 66(1): 121-134. Issue Date. 2015-08. URL. http://s-ir.sap.hokkyodai.ac.jp/dspace/handle/123456789/7849. Rights. Hokkaido University of Education.

(2) 北海道教育大学紀要（人文科学・社会科学編）第66巻第１号 Journal of Hokkaido University of Education（Humanities and Social Sciences）Vol. 66, No.1. 平成 27 年８月 August, 2015. 尊敬表現における名詞化について大賀京子北海道教育大学札幌校／国際交流・協力センター. Nominalisation in Subject Honorification OGA Kyoko International Centre, Sapporo Campus, Hokkaido University of Education. ABSTRACT It is widely accepted that in any language, there are ways to make speech polite. In Japanese, there are three types of honorific constructions for speakers to show their respect to: ⅰ the subject, ⅱ the object, or ⅲ the hearer of an utterance. In this study, I will focus on the two types of Subject Honorific (SH) and SHs appearing in the temporal sai adjunct phrase. Based on the observation that the combination of the honorific prefix o and the verb stem is nominal in that it can be followed by no, whereas it is verbal in that the preceding arguments can be associated with the nominative or accusative case markers, I argue that the SH in Japanese is an instance of‘mixed projections’and its nonargumental properties are derived from its internal structure.. １．Introduction It is widely accepted that in any language, there are ways to make speech polite. In Japanese, there are three types of honorific constructions for speakers to show their respect to: ⅰ the subject, ⅱ the object, or ⅲ the hearer of an utterance. According to Hasegawa (2006, p. 494), we will call them the ⅰ Subject Honorific (SH), ⅱ Object Honorific, and ⅲ Hearer Honorific, respectively. In the present study, I will focus on SHs. To illustrate their properties, let us first compare ⑴ and ⑵: ⑴. Tanaka-san-ga. Suzuki-san-o. Tanaka-Ms. -NOM Suzuki-Ms.-ACC. mat-ta. wait-PAST. ‘Ms. Tanaka waited for Ms. Suzuki.’. 121.

(3) 大賀京子. ⑵. Tanaka-san-ga. Suzuki-san-o. o-mati-ni nat-ta.. Tanaka-Ms.-NOM. Suzuki-Ms.-ACC HP-wait-HON-PAST. ‘Ms. Tanaka waited for Ms. Suzuki.’ (SH) In ⑴, the verb mat(u) ‘wait’ is in the plain past tense form, taking neither Tanaka-san ‘Ms. Tanaka’ nor Suzuki-san ‘Ms. Suzuki’ as someone the speaker shows respect. If the speaker knows that the subject Tanaka-san ‘Ms. Tanaka’ is a person for whom he/she should show respect, he/she needs to use the SH form as in. ⑵. In. ⑵, the stem of mat(u) ‘wait’ is sandwiched between two parts of an SH form: an. honorific prefix (henceforth HP) o and the past tense form of the predicate-like ni naru. The listener in ⑵ understands that the speaker respects the subject Tanaka-san ‘Ms. Tanaka’. Adding to the above mentioned SH, I will also discuss another type of SH, in which the verb stem is followed by the copula da as in ⑶: ⑶. Tanaka-san-ga. Suzuki-san-o. o-mati-da.. Tanaka-Ms.-NOM. Suzuki-Ms.-ACC HP-wait-COP. ‘Ms. Tanaka is waiting for Ms. Suzuki.’ (SH) In ⑶, the verb stem mat(u) is sandwiched between the prefix-like o and copula da. As another type of SH, ⑶ has the same meaning as ⑵. In this paper, I will focus on these two types of SH in Japanese and develop an analysis of the SH constructions. I will argue that the SH in Japanese is an instance of ‘mixed projections’ and its nonargumental properties are derived from its internal structure. This paper is organised as follows. In Section 2, I closely observe the two types of SH with respect to their nominal and nonargumental properties. In Section 3, I discuss properties of the particle ni that appears in SHs. In Section 4, I introduce my main proposal for the internal structure of SHs. In Section 5, I extend my analysis of SHs to other constructions including SHs. Section 6 concludes the discussion.. ２．Two types of SH and the SH in temporal sai adjunct phrases Let us now take a closer look at SHs. In both of the following examples, Tanaka-sensei ‘Tanaka-teacher’ is the person for whom the speaker shows respect. In ⑷, the stem of the verb kak(u) is sandwiched between the prefix-like o and predicate-like ni naru, while in ⑸, it is sandwiched between the same prefix o and copula da. ⑷ Tanaka-sensei-ga Tanaka-teacher-NOM. syosetu-o. o-kaki-ni naru.. novel-ACC. HP-write-HON.. ‘Professor Tanaka writes novels.’ ⑸ Tahaka-sensei-ga Tanaka-teacher-NOM. syosetu-o. o-kaki-da.. novel-ACC. HP-write-COP. ‘Professor Tanaka writes novels.’. 122.

(4) 尊敬表現における名詞化について. When ⑷ and ⑸ are followed by the time-denoting formal noun sai ‘occasion’, the sentences become temporal adjuncts (Shibatani, 1990, p. 247). When ⑷ is followed by sai, the predicate-like ni naru can be followed by sai directly, as in ⑹: ⑹. Tanaka-sensei-ga. syosetu-o. o-kaki-ni naru-sai,. Taro-wa. o-heya-ni. haira-nai.. Tanaka-teacher-NOM. novel-ACC. HP-write-HON-occasion. Taro-TOP. HP-room-into. enter-not. ‘When Professor Tanaka writes novels, Taro never enters into his room.’ When ⑸ is followed by the same formal noun sai to form a temporal adjunct phrase, on the other hand, the particle no is directly attached to o-kaki ‘HP-write’ and sai follows as in ⑺: ⑺. Tanaka-sensei-ga. syosetu-o. o-kaki-no-sai,. Taro-wa. o-heya-ni. haira-nai.. Tanaka-teacher-NOM. novel-ACC. HP-write-no-occasion. Taro-TOP HP-room-into enter-not. ‘When Professor Tanaka writes novels, Taro never enters his room.’ Given that the particle no can usually be attached to nominal expressions, we assume that ⑺ shows that the o-kaki ‘HP-write’ form is a sort of nominal expression. A question arises as to what the nominal property of o-kaki ‘HP-write’ stems from. To answer, let us compare SH constructions to normal sentences. Consider ⑻ below: ⑻ Taro-ga Taro-NOM. syosetu-o. kak-u.. novel-ACC. write-PRESENT. ‘Taro writes novels.’ In this sentence, there is no honorific expression; that is, Taro is not a person to whom the speaker should show his/her respect. When sai is added to ⑻, the construction is grammatical without the help of no as demonstrated by ⑼: ⑼. Taro-ga. syosetu-o. kak-u-sai,. Hanako-wa. heya-ni. haira-nai.. Taro-NOM. novel-ACC. write-PRESENT-occasion. Hanako-TOP room-into enter-not. ‘When Taro writes novels, Hanako never enters his room.’ However, the sentence becomes ungrammatical if the stem form of the verb kaki ‘write’ is associated with no directly before sai as in ⑽: ⑽. *Taro-ga. syosetu-o. kaki-no-sai,. Hanako-wa. heya-ni. haira-nai.. Taro-NOM. novel-ACC. write-no-occasion. Hanako-TOP. room-into. enter-not. ‘When Taro writes novels, Hanako never enters his room.’ Example ⑽ shows that the stem form of kaki does not have a nominal property and cannot be accompanied. 123.

(5) 大賀京子. by no. It is thus reasonable to suggest that the nominal property of o-kaki we observed in ⑺ originates in the affix o in SH constructions and it has the property of changing the following verb into a nominal expression. The combination of o and the verb stem (henceforth, o-V) is different from other usual nominal expressions, however. Let us compare an example of o-V, o-kaki ‘HP-write’, to the Sino-Japanese noun sippitu ‘writing’ in the sai adjunct phrase. As shown in ⑾, sippitu ‘writing’ and the light verb suru ‘do’ can form a complex verb corresponding to ‘write’: ⑾ Taro-ga Taro-NOM. syosetu-o. sippitu-suru.. novel-ACC. writing-do. ‘Taro writes novels.’ When sai is added to ⑾, it either follows directly as in ⑿, or the light verb of the sentence is replaced by no and both the nominative case marker ga and the accusative case marker o are replaced by no as in ⒀: ⑿. Taro-ga. syosetu-o. sippitu-suru-sai,. Taro-NOM. novel-ACC writing-do-occasion. Hanako-wa. heya-ni. haira-nai.. Hanako-TOP. room-into. enter-not. ‘When Taro writes novels, Hanako never enters his room.’ ⒀. Taro-no. syosetu-no. sippitu-no-sai,. Hanako-wa. heya-ni. haira-nai.. Taro-no. novel-no. writing-no-occasion. Hanako-TOP. room-into. enter-not. ‘On the occasion of Taro’s writing novels, Hanako never enters into his room.’ ⒀ shows that sippitu ‘writing’ itself is a noun and its nominal modifiers are associated with no (Kitagawa and Ross 1982, Suzuki 1989). Note, however, when o-V is followed by sai to form the SH temporal adjunct, no cannot replace the nominative and accusative case markers. Compare ⒁ to ⒂: ⒁. Tanaka-sensei-ga. syosetu-o. Tanaka-Professor-NOM novel-ACC. o-kaki-no-sai,. Taro-wa. o-heya-ni. HP-write-no-occasion. Taro-TOP HP-room-into. haira-nai. (= 7) enter-not. ‘When Professor Tanaka writes novels, Taro never enters his room.’ ⒂. *Tanaka-sensei-no. syosetu-no. o-kaki-no-sai,. Taro-wa. o-heya-ni. Tanaka-Professor-no. novel-no. HP-write-no-occasion. Taro-TOP HP-room-into. haira-nai. enter-not. ‘On the occasion of Professor Tanaka’s writing novels, Taro never enters his room.’ Comparing ⒁ to ⒂, we see that o-V has two different properties; it has some kind of nominal property in that it can be followed by no, whereas it retains its verbal properties in that its subject and object are associated with the nominative and accusative case markers (ga and o, respectively) in. ⒁ and these. cannot be replaced by no as seen in ⒂. Further, o-V in the SH is different from Sino-Japanese nouns in that it is not associated with case. 124.

(6) 尊敬表現における名詞化について. markers other than no. Compare ⒃ to ⒄: ⒃. a.. Taro-no. syosetu-no. sippitu-ga. hitobito-o. odorokase-ta.. Taro-no. novel-no. writing-NOM. people-ACC. surprise-PAST. ‘Taro’s writing of novels surprised people.’ b.. Watasi-wa. Taro-no. syosetu-no. sippitu-o. shira-nakat-ta.. I-TOP. Taro-no. novel-no. writing-ACC. know-not-PAST. ‘I didn’t know Taro’s writing of novels.’ ⒄. a.. *Tanaka-sensei-ga. syosetu-o. o-kaki-ga. hitobito-o. odorokase-ta.. Tanaka-teacher-NOM. novel-ACC. HP-write-NOM. people-ACC. surprise-PAST. o-kaki-o. shira-nak-atta.. HP-write-ACC. know-not-PAST. ‘Professor Tanaka’s writing novels surprised people.’ b.. *Watasi-wa. Tanaka-sensei-ga. syosetu-o. I-TOP. Tanaka-professor-NOM novel-ACC. ‘I didn’t know Professor Tanaka’s writing novels’ As shown in ⒃, sippitu ‘writing’ can be followed by the nominative and accusative case markers, whereas in ⒄, the o-V sequence o-kaki ‘HP-write’ cannot. The above observation shows that the properties of o-V are twofold. It is nominal in that it can be followed by no, whereas it is verbal in that the preceding arguments can be associated with the nominative or accusative case markers.. ３．The classification of ni in the SH Suzuki (1989, p. 374) discusses the structure of SHs focusing on case markers. He shows that the distribution of the particle ni in the SH parallels that of the ordinary dative case marker. Consider ⒅ and ⒆: ⒅ a.. Tanaka-sensei-ga. piano-o. o-hiki-ni naru.. Tanaka-Mr. NOM. piano-ACC HP-play-HON. ‘Mr. Tanaka plays the piano.’ b. ⒆ a.. *Tanaka-sensei-ga piano-o o-hiki-Ø naru. Tanaka-sensei-ga. isya-ni. naru. darou.. Tanaka-Mr.-NOM. doctor-ni. become. will. ‘Mr. Tanaka will become a doctor.’ b.. *Tanaka-sensei-ga isya-Ø naru darou.. 125.

(7) 大賀京子. The above paradigm shows that both ni in the SH and that of the dative case marker in ordinary sentences cannot drop. This leads Suzuki to claim that ni in SHs is a dative case marker and its preceding position can be case-marked as a noun phrase. Further, he suggests that the honorific is a noun phrase by showing the distribution of no as in ⒇-: ⒇. a.. [NP [CP sensei-ga. kaku] Ø [N hon]]. Teacher-NOM write. book. ‘The book the teacher writes’ b.. *[NP [CP sensei-ga. kaku]-no. Teacher-NOM write-GEN . a.. [NP [NP sensei]-no. [N hon]]. Teacher-GEN. book. [N hon]] book. ‘The teacher’s book’ b. a.. *[NP [NP sensei]Ø [N hon] [NP [α sensei-ga. o-kaki]-no. Teacher-NOM HP-write-GEN. [N hon] book. ‘The book the teacher writes’ (Honorific form) b.. *[NP [α sensei-ga o-kaki]Ø [N hon]. Examples ⒇- show that no is inserted if the constituent modifying the head N is an NP, whereas it cannot be inserted if the modifier is a relative clause as a CP. Based on the fact that no must be inserted in (22a), Suzuki claims that the category α is a DP, rather than a CP. Assuming that the category DP requires Case and the particle ni in SH is a dative case marker, Suzuki (1989, p. 380) proposes that in SH, the matrix verb naru ‘become’ takes a DP complement associated with the dative case marker ni, in which the head D takes an IP complement as illustrated in : . [IP [DP senseii]-ga [I’[VP [DP [IP PROi [I’ [VP [DP hon]-o Teacher-NOM. book-ACC. kak]. [I o -i[+N]]]] D]-ni. write HP [+N]. -DAT. naru] I]] become. ‘The teacher writes books.’ Suzuki claims that in , the verb kak ‘write’ raises to I or combines with a nominal ending realized as -i, turning the verb into a nominal. Based on Abney (1987), Suzuki suggests that the category DP requires Case and the o-V included in DP is thus associated with ni as a dative case marker. However, Suzuki’s analysis that o-V is included in the DP cannot account for the ungrammaticality of examples (17a) and (17b), in which the o-V is followed by nominal and accusative case markers,. 126.

(8) 尊敬表現における名詞化について. respectively. A question arises as to why o-V can be followed by the particle ni but not by nominative and accusative case markers if they are all case markers in Japanese. It is important to note that ni has been known as unique for appearing in various Japanese constructions with different syntactic and semantic functions. Takezawa (1993, p. 46) points out that ni has been classified as either ⅰ the infinitival form of da ‘be’ or ⅱ a postposition or dative case marker. The infinitival form of da can co-occur with adjectival nominatives and predicate nominals as in and (Takezawa, 1993, pp. 45-46): . Mary-ga. totemo. kirei-ni. natta.. Mary-NOM. very. pretty-ni. became. ‘Mary became very pretty.’ . John-ga. isya-ni. natta. John-NOM. doctor-ni. became. ‘John became a doctor.’ In , kirei ‘pretty’ is an adjectival noun followed by ni. Adjectival nouns are different from other normal nouns in that they cannot be associated with other typical case markers as illustrated by and : . *John-ga. kirei-o. John-NOM pretty-ACC. mita. saw. ‘John saw pretties.’ . *Kokoni. kirei-ga. aru.. here. pretty-NOM. be. ‘Here are pretties.’ In , isya ‘doctor’ appears as a predicate nominal, not as a referring expression. If isya appears with typical case markers as in and , the word refers to a person whose occupation is a doctor: . John-ga. isya-o. mita.. John-NOM. doctor-ACC. saw. ‘John saw a doctor.’ . Kokoni. isya-ga. iru.. here. doctor-NOM. be. ‘Here is a doctor.’ Given that o-V can be followed by ni and no, not by nominative or accusative case markers without the meaning change, I claim that the particle ni that follows o-V is an infinitival form of da (Hasegawa, 2006, p. 505) and no that connects the o-V and sai ‘occasion’ in ⑺ is the attributive form of da (Kuno, 1973, p.. 127.

(9) 大賀京子. 25). Therefore, the structure including o-V is different from that of other typical referring expressions in that o-V cannot be associated with the nominative and accusative case markers.. ４．The internal structure of o-V ni naru SH Having observed that nominative and accusative case markers cannot appear with o-V but the copula can in SHs, let us now consider the internal structure of the SH. Hasegawa (2006, pp. 506-508) proposes a structure adopting Suzuki’s analysis that the o-V part of the SH is a gerundive DP and Toribio’s (1990) analysis considering nar-u ‘become’ is a raising predicate. She assumes that the [+N] property of SH verbs is located at the head D and it is where the HP o resides. As for the connective morpheme -i before the copula, she suggests that it is a functional category, F, with the category feature [+N] that selects VP. Hasegawa’s structure is given in : . Sensei-ga. sinbun-o. o-yomi-ni nat-ta.. Teacher-NOM. newspaper-ACC. HP-read-HON-PAST. ‘The teacher read the newspaper.’ (honorific) V] v]] F]] D]] Cop]] V]] T]]. [TP [T’ [VP [V’ [CopP [Cop’[DP [[FP [F’ [vP DP [v’ [VP DP Sensei-ga. sinbun-o. yom. Teacher-NOM newspaper-ACC. -i. o. read. -ni nat -ta COP HON PAST. In her analysis, the HP o is D and requires sensei ‘teacher’ to move to the Spec of DP, checking off its agreement feature and further to the Spec of TP, checking off its Nominative Case feature. Hasegawa’s analysis of the HP o as D is based on the fact that noun expressions associated with another type of HP, go, refer to specific items that can be identified in a given discourse. Consider the examples and in which the nominal expressions that the HP go may accompany are marked with the accusative case marker (Hasegawa, 2006, pp. 508-509). . Sensei-ga. hisyo-ni. (go)-hon-o. o-watasi-ni nat-ta.. Teacher-NOM. secretary-DAT. HP-book-ACC. HP-pass-HON-PAST. ‘The teacher passed (his) book to the secretary.’ . Sensei-ga. (go)-hon-o. o-sagasi-ni nat-te i-ru.. Teacher-NOM. HP-book-ACC. HP-search-HON-PROG-PRESENT. ‘The teacher is looking for (his) book.’ Hasegawa points out that in and , the meaning of each sentence changes depending on whether or not the HP go is attached to hon ‘book’. Without go, hon ‘book’ would refer to any book, but with go, it would most probably refer to the one that the teacher wrote or owns. Having observed the meaning change, Hasegawa (2006, p. 509) claims that the HP go/o is D and has a definite (and/or specific) feature. 128.

(10) 尊敬表現における名詞化について. and it is carried over to the use of an HP in the SH form. She suggests that the presence of the HP in the SH indicates that the event described by the SH predicate refers to a definite/specific event and thus the definiteness/specificity of the sentence stems from the HP as D. However, as we observed in (17a) and (17b), o-V cannot be followed by nominative or accusative case markers. If o-V were included in the DP, we would not be able to explain why nominative and accusative case markers, which usually have no problem being attached to typical arguments, cannot follow o-V. Given that da, its infinitival form ni and its attributive form no can appear with non-referential nominal expressions, the structure that o-V is included in is not accounted for as a DP (Longobardi, 1994). According to Ivana & Sakai (2007, p. 184), I claim that the HP o is a nominalising honorific prefix that resides on, not D, but another nominal functional category, Honorification (henceforth, Hon). Based on Borsley & Kornfilt’s (2000, p. 102) analysis for mixed projections, I propose a structure in which T, as a verbal functional category, appears above a vP and then Hon as a nominal functional category above TP. When the o-V is followed by ni naru, the structure is as illustrated in : . Tanaka-sensei-ga. syosetu-o. o-kaki-ni naru.. Tanaka-teacher-NOM. novel-ACC. HP-write-HON. ‘Professor Tanaka writes novels.’. 129.

(11) 大賀京子. I assume that the verb kak ‘write’ is included in a VP shell structure. Harada (1976, p. 517) claims that the verb is infinitivized if the prefix o is adjoined to it. In the present analysis, I claim that Harada’s ‘infinitivization’ process is brought about by the infinitival T realised as the morpheme -i, taking vP as its complement.1 The infinitival TP is a complement of Hon where o resides, and the HonP is a complement of ni as Cop, that is the infinitival form of the copula. The CopP is the complement of the verb naru as a light verb (Grimshaw & Mester 1988, Ivana & Sakai 2007).2 Given that Japanese is a head-final language, I put the head Hon on the right side, although o appears on the left side of the verb in the actual order. Based on Lasnik (1999) and Ivana & Sakai (2007), I assume that the overt morpheme order is obtained by the morphological merger operation at PF and through the operation, o is attached to the left side of the infinitivized verb. The present analysis is different from other previous works. Ivana & Sakai (2007, p. 184) claim that o as the head H (Hon in the present study) takes an NP whose head N is the morpheme -i as a renyokei suffix. However, as Hasegawa (2006, p. 507) points out, the connective form that ends with -i is not always a nominal. Let us consider the examples of conjoined SHs in (Hasegawa, 2006, pp. 505 -506): . a.. Tanaka-sensei-wa. otya-o. Tanaka-teacher-TOP tea-ACC. o-nomi-ni,. Kato-sensei-wa. coffee-o. o-nomi-ni nat-ta.. HP-drink-HON. Kato-teacher-TOP. coffee-ACC HP-drink-HON-PAST. ‘Prof. Tanaka drank tea and Prof. Kato drank coffee.’ b.. *Tanaka-sensei-wa otya-o o-nomi, Kato-sensei-wa coffee-o o-nomi-ni nat-ta.. As (34a) shows, the copula of the first conjunct turns into the connective form ni in conjoined SH sentences in Japanese. The umgrammaticality of (34b) shows that o-nomi ‘HP-drink’ cannot serve the function of conjoining sentences. This is parallel to predicative nominals as shown in : . a.. Taro-ga. sensei-ni,. Hanako-ga. isya-ni. nat-ta.. Taro-NOM. teacher-COP. Hanako-NOM. doctor-COP. become-PAST. ‘Taro became a teacher and Hanako became a doctor.’. 1 The infinite T is not always realised as the morpheme -i. There is a group of verbs whose infinitival form has the morpheme -e as below: ⅰ. Tanaka-sensei-ga. mado-o. o-sime-ni nat-ta.. Tanaka-teacher-NOM. window-ACC. HP-close-HON-PAST. ‘Professor Tanaka closed the window.’ 2 The subject Tanaka-sensei-ga ‘Tanaka-teacher-NOM’ may move from within the vP to the Spec of the embedded TP where its Nominative Case feature is checked and further moves to the Spec of HonP where its honorific agreement feature is checked, according to Hasegawa’s (2006) overt movement analysis. In this paper, I will not discuss how the Case and agreement features are checked.. 130.

(12) 尊敬表現における名詞化について. b.. ??Taro-ga sensei, Tanako-ga isya-ni nat-ta.. In (35a), the copula accompanying the predicate nominal in the first conjunct turns into the connective form, and if it is deleted as in (35b), the grammaticality of the sentence is degraded. and seem to show that o-V in and a predicative noun in are both nominal and neither of them can function as a sentential conjunction without ni. Note, however, that the verb attached with -i attached is not always nominal. Consider below: . Tanaka-sensei-wa. otya-o. nomi,. Kato-sensei-wa. coffee-o. non-da. (non-honorific). Tanaka-teacher-TOP. tea-ACC. drink. Kato-teacher-TOP. coffee-ACC. drink-PAST. ‘Professor Tanaka drank tea and Professor Kato drank coffee.’ In a non-honorific sentence such as , the verb of the first conjunct is in the connective form nomi, and the conjoined sentence is grammatical. This suggests that the connective form itself is not always a nominal. That is, the nominal property of o-V in SH is not accounted for as originating from -i, as opposed to Ivana & Sakai’s analysis. In my analysis, the nominal properties of the o-V originated from Hon as a nominal functional category. Based on Longobardi’s (1994) claim that DP functions as an argument and the fact that typical arguments can be associated with case markers such as nominative and accusative case markers in Japanese, I suggest that the o-V in SHs that is unable to be accompanied by any typical case markers is part of a ‘nonargumental’ nominal whose head is Hon as a functional category that takes the infinitival TP as its complement. Hon is thus different from D in that it and TP form what is equivalent to a predicate nominal, and Hon lacks a feature of turning the infinitival TP into an argument. Having both Hon as a nominal functional category and TP as a verbal projection, the proposed HonP is accounted for as an instance of mixed projections consisting of a nominal and a verbal/clausal part (Borsley & Kornfilt, 2000; Panagiotidis, 2010).3. ５．Structures of the SH in temporal sai adjunct phrases and of o-V da SH As shown in ⑺, when o-V is within a temporal sai adjunct phrase, the subject within the phrase is accompanied with the nominative case marker. This suggests that the Nominative Case must be checked off within the adjunct phrase without the help of T of the matrix sentence. I thus propose that SH in the temporal sai adjunct phrase has the same CopP structure as SH does in . The proposed structure for the temporal sai adjunct phrase is shown in : 3 The so-called gerunds have been familiar cases of mixed projections in the literature. According to Panagiotidis (2010), gerunds in Hebrew and Greek cannot be arguments, whereas gerunds in English can appear in argument positions. Hebrew gerunds can only be found in the complement of a temporal preposition, and Greek gerunds can only function as adverbial adjuncts. Panagiotidis claims that Hebrew and Greek gerunds are embedded within a temporal PP headed by the null temporal head P. Although it is an interesting similarity that Hebrew and Greek gerunds and HonP in the present study do not appear in argument positions, I will leave the issue for future research.. 131.

(13) 大賀京子. . Tanaka-sensei-ga. syosetu-o. o-kaki-no-sai,. (Taro-wa. Tanaka-teacher-NOM. novel-ACC. HP-write-no-occasion (Taro-TOP. o-heya-ni. haira-nai).(=7). HP-room-into. enter-not). ‘When Professor Tanaka writes novels, (Taro never enters his room.)’. In , the temporal formal noun sai ‘occasion’ takes the CopP whose head is no, the attributive form of the copula, whereas in , the light verb naru takes the CopP whose head is ni, the infinitive form of the copula. I claim that the temporal sai adjunct phrase is an NP and functions as a bare NP adverb.4 I would like to claim that the above proposal can be extended to the other type of SH, o-V-da type, repeated below as : . Tahaka-sensei-ga. syosetu-o. o-kaki-da. (=5). Tanaka-teacher-NOM. novel-ACC. HP-write-COP. ‘Professor Tanaka writes novels.’ I claim that this type of SH sentence includes the same CopP structure as the other SH type, and the difference is that the CopP is a complement to the matrix T as illustrated below: 4 In this study, I will not discuss the detail of how the Nominative Case is licensed.. 132.

(14) 尊敬表現における名詞化について. . In summary, I have claimed that the nominal property of o-V originates from the existence of an honorific nominal functional category, Hon, which is different from D in that it cannot serve to form arguments. The HonP headed by Hon is thus equivalent not to an argumental DP but to a predicate nominal and can be a complement to the Cop. Due to its nonargumental properties, HonP needs to be merged with Cop to become a predicate as illustrated in and , or a nominal modifier as illustrated in .. ６．Conclusion In this paper, I have argued that SH in Japanese includes nominalisation by a nominal functional category Hon, which has the property of taking a TP to form a nominal, but not argumental, expression. Having both Hon as a nominal functional category and TP as a verbal projection, the proposed HonP is accounted for as an instance of mixed projections consisting of a nominal and a verbal/clausal part. Due to its nonargumental properties, HonP needs to be merged with Cop to become a predicate or a modifier.. REFERENCES Abney, S. (1987). The English noun phrase in its sentential aspect, Ph.D. dissertation, MIT, Cambridge, MA. Borsley, R. D. & Kornfilt, J. (2000). Mixed extended projections. In R. D. Borsley (Ed.), The nature and function of syntactic. 133.

(15) 大賀京子. categories (Syntax and Semantics, 32) (pp. 101-131). New York: Academic Press. Grimshaw, J. & Mester, A. (1988). Light verbs and theta marking. Linguistic Inquiry, 19, 205-232. Harada, S. I. (1976). Honorifics, In Shibatani, M. (Ed.), Japanese Generative Grammar (Syntax and Semantics, 5) (pp. 499561). New York: Academic Press. Hasegawa, N. (2006). Honorifics. In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax, vol. II, (pp. 493-543). Oxford: Blackwell. Ivana, A. & Sakai, H. (2007). Honorification and light verbs in Japanese. Journal of East Asian Linguistics, 16, 171-191. Kitagawa, C. & Ross, C. (1982). Prenominal modification in Chinese and Japanese. Linguistic Analysis, 9, 19-53. Kuno, S. (1973). The structure of the Japanese language. Cambridge: MIT Press. Lasnik, H. (1999). Minimalist analysis. Oxford: Blackwell. Longobardi, G. (1994). Reference and proper names: a theory of N-movement in syntax and logical form. Linguistic Inquiry, 25, 609-665. Panagiotidis, E. P. (2010). Nonargumental mixed projections. Syntax, 13, 165-182. Shibatani, M. (1990). The languages of Japan. Cambridge: Cambridge University Press. Suzuki, T. (1989). A syntactic analysis of an honorific construction o…-ni naru under the DP hypothesis: toward a unified theory of honorification. In E. J. Fee & K. Hunt (Eds.), The proceedings of the 8th West Coast conference on formal linguistics (pp. 373-383). Stanford Linguistics Association, CSLI, Stanford University. Takezawa, K. (1993). Secondary predication and locative/goal phrases. In N. Hasegawa (Ed.), Japanese syntax in comparative grammar (pp. 45-77). Tokyo: Kurosio Publishers. Toribio, A. J. (1990). Specifier-head agreement in Japanese. In A. L. Halpern (Ed.), The proceedings of the 9th West Coast conference on formal linguistics (pp. 535-548). Stanford Linguistics Association, CSLI, Stanford University.. （札幌校准教授）. 134.

(16)