| Section 9 | ||
|---|---|---|
 |
 |
|
| Section 9 | ||
|---|---|---|
| |
|
|
The predicate is the most important syntactic unit in a clause, with the functions of almost all other constituents being oriented thereto. Predicates in Japanese can be extended by elements that specify voice, aspect, polarity, modality, evidentiality. The basic elements of the predicate are discussed below.
This section discusses the annotation of verbs. Expressions that are treated as lexical compound verbs in the BCCWJ and CSJ corpora (e.g., かきわける, そぎとる, とりいれる, はりつける, etc.) are kept as one segment in the Keyaki Treebank, but auxiliary verbs of various types (which are chunked in LUWs in BCCWJ and CSJ) are separated into their constituent segments in the Keyaki Treebank and labelled VB2. A traditional definition of “auxiliary verb” is a form that can appear as a core lexical verb in some contexts, but is also used in secondary verb position to express a value in a grammatical category such as deixis, aspect, polarity, etc. (See section 9.1.1 for details). We present simple instances of core verbal predicates (VB) first.
When appearing with subjects, predicates formed on VB project clauses:
(103)
ほかにも立候補の動きがある。
( (IP-MAT (PP (NP (N ほか))
(P に)
(P も))
(PP (NP (PP (NP (N 立候補))
(P の))
(N 動き))
(P が))
(NP-SBJ *が*)
(VB ある)
(PU 。))
(ID 100_news_KAHOKU_28))
(104)
芳一は大きな門口に達したのだと覚った――
( (IP-MAT (PP (NP (NPR 芳一))
(P は))
(NP-SBJ *)
(CP-THT (IP-SUB (NP-SBJ *pro*)
(PP (NP (IP-REL (NP-SBJ *T*)
(ADJN 大きな)
(AX *))
(N 門口))
(P に))
(VB 達し)
(AXD た)
(FN の)
(AX だ))
(P と))
(VB 覚っ)
(AXD た)
(PU ――))
(ID 59_aozora_Togawa-1937))
(105)
午まになったらまた来るがら。」
( (CP-FINAL (IP-SUB (NP-SBJ *speaker*)
(IP-ADV (PP (NP (N 午ま))
(P に))
(VB なっ)
(P たら))
(CND *)
(ADVP (ADV また))
(VB 来る))
(P がら)
(PU 。)
(-RRB- 」))
(ID 363_aozora_Miyazawa-1934))
(106)
家事一切に関わらず、のんびりと心のぜいたくをする。
( (IP-MAT (NP-SBJ *speaker*)
(IP-ADV (PP (NP (N 家事)
(Q 一切))
(P に))
(VB 関わら)
(NEG ず))
(CONJ *)
(PU 、)
(PP (ADVP (ADV のんびり))
(P と))
(PP (NP (PP (NP (N 心))
(P の))
(N ぜいたく))
(P を))
(NP-OB1 *を*)
(VB する)
(PU 。))
(ID 15_news_KAHOKU_10414))
(107)
最近の親は、子供たちに自由にいろいろな所に行かせる。
‘Contemporary parents let their children go to various places freely .’
( (IP-MAT (PP (NP (PP (NP (N 最近))
(P の))
(N 親))
(P は))
(NP-SBJ *)
(PU 、)
(PP (NP (N 子供たち))
(P に))
(NP-OB1 *に*)
(IP-SMC (ADVP (ADJN 自由)
(AX に))
(PP (NP (IP-REL (NP-SBJ *T*)
(ADJN いろいろ)
(AX な))
(N 所))
(P に))
(VB 行か))
(VB せる)
(PU 。))
(ID 673_textbook_purple_intermediate))
(108)
見知らぬ人は言葉をやわらげて言い出した、
( (IP-MAT (PP (NP (IP-REL (NP-OB1 *T*)
(NP-SBJ *pro*)
(VB 見知ら)
(NEG ぬ))
(N 人))
(P は))
(NP-SBJ *)
(IP-ADV (PP (NP (N 言葉))
(P を))
(NP-OB1 *を*)
(VB やわらげ)
(P て))
(CONJ *)
(VB 言い)
(VB2 出し)
(AXD た)
(PU 、))
(ID 46_aozora_Togawa-1937))
(109)
ただ餅を搗く音だけする。
( (IP-MAT (ADVP (ADV ただ))
(PP (NP (IP-EMB (NP-SBJ *pro*)
(PP (NP (N 餅))
(P を))
(NP-OB1 *を*)
(VB 搗く))
(N 音))
(P だけ))
(NP-SBJ *)
(VB する)
(PU 。))
(ID 504_aozora_Natsume-1908))
The class of verbs is divided sub-classes:
As mentioned in section 5, VB, VB0, and VB2 are immediately dominated by the IP, to form a relatively flat structure. VB is the first element in a verbal predicate. In a complex verbal syntagm VB will typically have the form of a stem. The most common VBs, presented here in their citation forms, are する, ある, なる, 見る, 行く, 思う, 言う. But these and other VBs frequently appear in some form of combining stem, in which case they are followed by either a VB2, or some other verbal morphology such as passive PASS, negative NEG, past tense AXD, etc. As special cases of VB, the causative verbal auxiliary and the indirect passive verbal auxiliary appear immediately after small clauses that contain a core verbal predicate. As the first inflecting elements in an IP, they technically satisfy the definition for VB (discussed in more detail below).
VB0s are the stems or inflected forms of verbs like する, いたす, なさる, 申し上げる, 下さる, できる, and the suffix がる. VB0s appear after VBs that are sometimes called “verbal nouns” such as お願い, 移動, チェック, むかむか, etc. They too can be followed by VB2, PASS, NEG, and the like.
(110)
「やあ耕助君、失敬したねえ。」
( (CP-FINAL (-LRB- 「)
(IP-SUB (INTJ やあ)
(NP-SBJ (NPR 耕助君))
(PU 、)
(VB 失敬)
(VB0 し)
(AXD た))
(P ねえ)
(PU 。)
(-RRB- 」))
(ID 659_aozora_Miyazawa-1934))
VB2s are non-initial verbs that follow verbs that normally can inflect. When a VB2 immediately follows a VB it is frequently of a class of auxiliary verbs that includes aspectual verbs 始める, 出す, かける, 続ける, 終える, 終る, 止む; potential verbs 兼る, 得る, 抜く, 損う; degree verbs 過ぎる, 尽す, 果てる, 切る, 足りる; direction verbs 込む, 回る, 去る; and various other verbs 忘れる, 間違える, 直す, 合う, 立てる, 付ける; etc. Members of this class can combine with other members.
(111)
震災は人々の記憶から薄れ始めているのか。
( (CP-QUE (IP-SUB (PP (NP (N 震災))
(P は))
(NP-SBJ *)
(PP (NP (PP (NP (N 人々))
(P の))
(N 記憶))
(P から))
(VB 薄れ)
(VB2 始め)
(P て)
(VB2 いる)
(FN の))
(P か)
(PU 。))
(ID 6_news_KAHOKU_39))
When, under the same IP, a VB2 immediately follows (P て), it is of a different class of auxiliary verbs that can also mark aspect or other meanings: いる, ある, おく, しまう, みる, みせる, あげる, くれる, やる, くださる, いく，くる, いらっしゃる, まいる, お出で, ご覧, 頂戴, etc. Members of this class can combine with other members.
(112)
今、調べています。
‘I am checking it now.’
( (IP-MAT (NP-SBJ *pro*)
(NP-OB1 *pro*)
(NP-TMP (N 今))
(PU 、)
(VB 調べ)
(P て)
(VB2 い)
(AX ます)
(PU 。))
(ID 478_textbook_kisonihongo))
When verbs する (or analogous forms いたす, なさる, etc.) and なる follow either a く-form い-adjective or [nominal predicate + に], frequently a small clause is projected by the predicate preceding する or なる making する or なる the first verb in a verbal syantagm, hence VB. There are three conditions that a small clause must satisfy in this context: (i) the head of the small clause must denote a property, not an event; (ii) NP-OB1 of する or the NP-SBJ of なる must be interpretable as the subject of the head of the small clause; (iii) NP-OB1 of する or the NP-SBJ of なる must precede the small clause, unless the small clause is marked with a toritate particle. If these conditions are satisfied, then a small clause analysis is adopted. To instantiate, in first clause in the example below, the realisation of the action denoted by 細帯を長くして entails that 長い is predicated of 細帯. Accordingly い-adjective 長く projects a small clause IP-SMC here.
(113)
そうして細帯を長くして、子供を縛っておいて、その片端を拝殿の欄干に括りつける。
( (IP-MAT (NP-SBJ *pro*)
(CONJ そうして)
(IP-ADV (PP (NP (N 細帯))
(P を))
(NP-OB1 *を*)
(IP-SMC (ADJI 長く))
(VB し)
(P て))
(CONJ *)
(PU 、)
(IP-ADV (PP (NP (N 子供))
(P を))
(NP-OB1 *を*)
(VB 縛っ)
(P て)
(VB2 おい)
(P て))
(CONJ *)
(PU 、)
(PP (NP (D その)
(N 片端))
(P を))
(NP-OB1 *を*)
(PP (NP (PP (NP (N 拝殿))
(P の))
(N 欄干))
(P に))
(VB 括り)
(VB2 つける)
(PU 。))
(ID 565_aozora_Natsume-1908))
In the example below, the realisation of the event denoted by 路はだんだん暗くなる entails that 暗い is predicated of 路, so 暗く projects a small clause IP-SMC.
(114)
路はだんだん暗くなる。
( (IP-MAT (PP (NP (N 路))
(P は))
(NP-SBJ *)
(ADVP (ADV だんだん))
(IP-SMC (ADJI 暗く))
(VB なる)
(PU 。))
(ID 209_aozora_Natsume-1908))
As noted above, the analysis shown for infinitive い-adjectives in these contexts is extended to infinitive ADJN AX and infinitive NP-PRD AX.
(115)
仕事はあっても利益にならず、豊作貧乏の状況だ。
( (IP-MAT (NP-SBJ *pro*)
(IP-ADV (PP (NP (N 仕事))
(P は))
(NP-SBJ *)
(PP (IP-ADV (VB あっ)
(P て))
(P も))
(CONJ *)
(IP-SMC (NP-PRD (N 利益))
(AX に))
(VB なら)
(NEG ず))
(SCON *)
(PU 、)
(NP-PRD (PP (NP (N 豊作貧乏))
(P の))
(N 状況))
(AX だ)
(PU 。))
(ID 53_news_KAHOKU_113))
Occasionally a null NP-SBJ2 needs to be interpolated as a controller for the small clause in these constructions:
(116)
ソニーが新しい社長になった
( (IP-MAT (PP (NP (NPR ソニー))
(P が))
(NP-SBJ *が*)
(IP-SMC (NP-PRD (IP-REL (NP-SBJ *T*)
(ADJI 新しい))
(N 社長))
(AX に))
(VB なっ)
(AXD た))
(ID 48_misc_EXAMPLE))
This small-clause analysis is contrasted with constructions such as those in the examples below, where the subject-predicate relation doesn't hold (as the matrix subject controls a position further down in the structure) and the condition on property ascription isn't met.
(117)
今後の医療運営は被災地の人口流出にも向き合うことになる。
( (IP-MAT (PP (NP (PP (NP (N 今後))
(P の))
(N 医療運営))
(P は))
(NP-SBJ *)
(PP (NP (IP-EMB (PP (NP (PP (NP (N 被災地))
(P の))
(N 人口流出))
(P に)
(P も))
(VB 向き合う))
(N こと))
(P に))
(NP-OB1 *に*)
(VB なる)
(PU 。))
(ID 25_news_KAHOKU_93))
(118)
「昨年の今ごろ」のニュースや話題が閲覧できるようになる。
( (IP-MAT (NP-SBJ *pro*)
(PP (NP (IP-EMB (PP (NP (PP (NP (-LRB- 「)
(PP (NP (N 昨年))
(P の))
(N 今ごろ)
(-RRB- 」))
(P の))
(NML (CONJP (NP (N ニュース))
(P や))
(NP (N 話題))))
(P が))
(NP-OB1 *が*)
(VB 閲覧)
(VB0 できる))
(N よう))
(P に))
(VB なる)
(PU 。))
(ID 30_news_KAHOKU_97))
する (and analogous forms) and なる appear as VB2 in honorific and humilific verbal syntagms. Honorific forms of verbs are variably prefixed by お or ご (御) and take the form of verbal nouns. In such cases, the verbal syntagm is composed as follows: As the first verb in a verbal syntagm, an honorific or humilific form is marked VB. The honorific form is immediately followed by (P に) and then (VB2 なる). The humilific form is immediately followed by (VB2 する) or analogous verb 致す.
(119)
鈴木先生は本をお書きになった。
‘Mr. Suzuki wrote a book.’
( (IP-MAT (PP (NP (NPR 鈴木先生))
(P は))
(NP-SBJ *)
(PP (NP (N 本))
(P を))
(NP-OB1 *を*)
(VB お書き)
(P に)
(VB2 なっ)
(AXD た)
(PU 。))
(ID 1295_textbook_kisonihongo))
(120)
花子は鈴木先生に結果をご報告した。
‘Hanako reported to Mr. Suzuki the result.’
( (IP-MAT (PP (NP (NPR 花子))
(P は))
(NP-SBJ *)
(PP (NP (NPR 鈴木先生))
(P に))
(NP-OB2 *に*)
(PP (NP (N 結果))
(P を))
(NP-OB1 *を*)
(VB ご報告)
(VB0 し)
(AXD た)
(PU 。))
(ID 1311_textbook_kisonihongo))
Some valence-increasing constructions require a small clause analysis. This affects some instances donative verbs もらう and いただく immediately following (VB ...) (P て). They can be analyzed as subcategorising for a に-marked NP-OB1 which controls a subject position in a small clause IP-SMC. The IP-SMC contains (VB ...) (P て) and any arguments or modifiers thereof. Thus もらう is the first verb VB in the matrix verbal syntagm in such a context, as is seen in the example below.
(121)
私は鈴木さんに答えを教えてもらった。
‘I got Mr. Suzuki to teach me the answer.’
( (IP-MAT (PP (NP (PRO 私))
(P は))
(NP-SBJ *)
(PP (NP (NPR 鈴木さん))
(P に))
(NP-OB1 *に*)
(IP-SMC (PP (NP (N 答え))
(P を))
(NP-OB1 *を*)
(VB 教え)
(P て))
(VB もらっ)
(AXD た)
(PU 。))
(ID 299_textbook_kisonihongo))
The indirect passive construction is also a valence-increasing construction, one where an affected argument is added as NP-SBJ to a construction with only one core lexical verb. The logical subject of the core verbal expression is typically marked with に, but can be shown to have moved to a position sister to an added argument which is the (affected) matrix subject (e.g., in the sentence 私は嫌いな人に不本意に自分の書いた論文を褒められてしまった). From that position the logical subject controls into a subordinate small clause (which typically includes any other arguments or modifiers of the core verbal expression). This bi-clausal analysis accounts for the increase in valence along with some syntactic other phenomena. But this analysis requires a bound morpheme (VB (ら)れる) to immediately follow IP-SMC, becoming a core predicate of the matrix IP. See section 30.10. Note that the direct passive and the use of (ら)れる in direct passive, potential, spontaneous, and honorific constructions do not trigger the bi-clausal analysis. (ら)れる in direct passive is tagged PASS, and (ら)れる in potential, spontaneous, and honorific constructs is tagged VB2.
(122)
私は弟にケーキを食べられた。
( (IP-MAT (PP (NP (PRO 私))
(P は))
(NP-SBJ *)
(PP (NP (N 弟))
(P に))
(NP-OB1 *に*)
(IP-SMC (PP (NP (N ケーキ))
(P を))
(NP-OB1 *を*)
(VB 食べ))
(VB られ)
(AXD た)
(PU 。))
(ID 17_misc_EXAMPLE))
The causative construction is also a valence-increasing construction. Here too the logical subject of the core verbal expression is typically marked with に (but を is possible in some contexts), and can be shown to have moved to a position sister to an added argument which is the matrix subject. This bi-clausal analysis accounts for the increase in valence along with some syntactic other phenomena. But this analysis requires a bound morpheme (VB (さ)せる) to immediately follow IP-SMC, becoming the first verbal element of the matrix IP, hence a VB. See section 30.8.
(123)
私は弟にケーキを食べさせた。
( (IP-MAT (PP (NP (PRO 私))
(P は))
(NP-SBJ *)
(PP (NP (N 弟))
(P に))
(NP-OB1 *に*)
(IP-SMC (PP (NP (N ケーキ))
(P を))
(NP-OB1 *を*)
(VB 食べ))
(VB させ)
(AXD た)
(PU 。))
(ID 15_misc_EXAMPLE))
Note that a predicate may appear to form a string in the case of the “purpose of motion construction,” but this is analyzed as a separate constituent: Verbs of motion can be modified by constituents that end with an infinitive form of a verb followed by に. The infinitive verb can take its own arguments, distinct from those selected by the verb that follows, so the infinitive verb is analyzed as projecting a clause marked by a particle に. The PP-PRP containing the purpose of motion clause is simply adverbial to the verb of motion, with little restriction on its order of appearance in the sentence. See section 30.13.
(124)
しばらくして、若者２人が助けに来ました。
( (IP-MAT (ADVP (ADV しばらくして))
(PU 、)
(PP (NP (N 若者)
(NUMCLP (NUM ２人)))
(P が))
(NP-SBJ *が*)
(PP-PRP (IP-NML (NP-OB1 *speaker*)
(VB 助け))
(P に))
(VB 来)
(AX まし)
(AXD た)
(PU 。))
(ID 105_news_KAHOKU_34))
Finally, there is a bound suffix がる that inflects like a verb, but as it is a bound suffix, it is given the category of AX.
ADJIs (also called い-adjectives) such as 大きい and うつくしい have a unique conjugation paradigm which, for instance, lacks the imperative inflection. ADJIs have some limitations on co-occurrence with modifiers and predicate extensions, and in certain contexts they exhibit special focus interactions with particles. But with regard to projection, they behave as other predicates. When appearing with subjects, outside of a noun phrase, they can project a main clause, IP-SUB under CP, IP-ADV, or IP-SMC, with basically the same constraints on grammatical function and grammatical marking as verbs and other predicates.
(125)
新市を軌道に乗せる意欲は強い。
( (IP-MAT (PP (NP (IP-EMB (NP-SBJ *pro*)
(PP (NP (N 新市))
(P を))
(NP-OB1 *を*)
(PP (NP (N 軌道))
(P に))
(VB 乗せる))
(N 意欲))
(P は))
(NP-SBJ *)
(ADJI 強い)
(PU 。))
(ID 160_news_KAHOKU_28))
(126)
そんなこともないですか？
( (CP-QUE (IP-SUB (PP (NP (D そんな)
(N こと))
(P も))
(NP-SBJ *)
(ADJI ない)
(AX です))
(P か)
(PU ？))
(ID 21_news_KAHOKU_10414))
(127)
人間で言えば８０歳近いのに、まだやんちゃですよ」
( (CP-FINAL (IP-SUB (NP-SBJ *pro*)
(PP (IP-ADV (IP-ADV (NP-SBJ *speaker*)
(NP-OB1 *pro*)
(PP (NP (N 人間))
(P で))
(VB 言え)
(P ば))
(CND *)
(PP (NP (NUMCLP (NUM ８０)
(CL 歳)))
(P *に*))
(ADJI 近い))
(P のに))
(SCON *)
(PU 、)
(ADVP (ADV まだ))
(ADJN やんちゃ)
(AX です))
(P よ)
(-RRB- 」))
(ID 8_news_KAHOKU_13153))
(128)
家族愛の深さは尊いが、互いを心配するあまり、津波避難が遅れる場合もある。
( (IP-MAT (PP (IP-ADV (PP (NP (PP (NP (N 家族愛))
(P の))
(N 深さ))
(P は))
(NP-SBJ *)
(ADJI 尊い))
(P が))
(CONJ *)
(PU 、)
(NP-ADV (IP-EMB (NP-SBJ *pro*)
(PP (NP (N 互い))
(P を))
(NP-OB1 *を*)
(VB 心配)
(VB0 する))
(N あまり))
(PU 、)
(PP (NP (IP-EMB (PP (NP (N 津波避難))
(P が))
(NP-SBJ *が*)
(VB 遅れる))
(N 場合))
(P も))
(NP-SBJ *)
(VB ある)
(PU 。))
(ID 45_news_KAHOKU_63))
(129)
このいちごは甘くておいしいです。
( (IP-MAT (PP (NP (D この)
(N いちご))
(P は))
(NP-SBJ *)
(IP-ADV (ADJI 甘く)
(P て))
(CONJ *)
(ADJI おいしい)
(AX です)
(PU 。))
(ID 27_misc_BUFFALO))
(130)
阪神大震災でも心の問題は３年目に多くなった。
( (IP-MAT (PP (NP (NPR 阪神大震災))
(P で)
(P も))
(PP (NP (PP (NP (N 心))
(P の))
(N 問題))
(P は))
(NP-SBJ *)
(PP (NP (NUMCLP (NUM ３)
(CL 年目)))
(P に))
(IP-SMC (ADJI 多く))
(VB なっ)
(AXD た)
(PU 。))
(ID 25_news_KAHOKU_616))
When an い-adjective appears with a subject (overt or null) and modifies a noun, it projects a relative clause (IP-REL) or a content/specificational clause (IP-EMB).
(131)
――ある早朝のこと――はげしい雨がガラス窓を打っていた。
( (IP-MAT (NP-TMP (-LRB- ――)
(PP (NP (D ある)
(N 早朝))
(P の))
(N こと)
(-RRB- ――))
(PP (NP (IP-REL (NP-SBJ *T*)
(ADJI はげしい))
(N 雨))
(P が))
(NP-SBJ *が*)
(PP (NP (N ガラス窓))
(P を))
(NP-OB1 *を*)
(VB 打っ)
(P て)
(VB2 い)
(AXD た)
(PU 。))
(ID 918_aozora_Harada-1960))
(132)
おそらくすでに春が近いしるしだろう
( (IP-MAT (NP-SBJ *pro*)
(ADVP (ADV おそらく))
(NP-PRD (IP-EMB (ADVP (ADV すでに))
(PP (NP (N 春))
(P が))
(NP-SBJ *が*)
(ADJI 近い))
(N しるし))
(AX だろ)
(MD う))
(ID 919_aozora_Harada-1960))
There are many い-adjectives having specific senses that require a second argument. A partial list of 2-place い-adjectives, together with the particle(s) that typically mark(s) the non-subject argument:
Note how the い-adjective 欲しい etc., takes two argument NPs, both of which can be marked with が. The second one is assigned the role OB1. A careful distinction must be made between this kind of sentence and the double subject sentence. If either of the が-marked NPs, to the exclusion of the other, can combine with the predicate to build up an appropriate sentential meaning with a context, they are either the subject or direct object of the same predicate. For example, suppose that a sentence 私がチョコレートが欲しい is given, both of the sentences below are semantically well-formed and can stand alone with a context.
In this case, 私が is the subject and チョコレートが is the direct object of the sentence.
By contrast, given a sentence 象が鼻が長い (The elephant-SBJ the nose-SBJ2 is long), since in the sequence 象が長い the NP 象が cannot be interpreted as the subject of the い-adjective 長い while preserving the meaning of the the original sentence (because it is not the elephant that is long), the whole is a double-subject sentence with 象が as SBJ; the second subject 鼻が is assigned the role SBJ2.
Auxiliary い-adjective 欲しい takes an IP-SMC when following (P て):
(133)
あさってまでに完成してほしい。
‘I would like you to finish it by the day after tomorrow.’
( (IP-MAT (NP-SBJ *speaker*)
(NP-OB1 *pro*)
(IP-SMC (PP (NP (N あさって))
(P まで)
(P に))
(VB 完成)
(VB0 し)
(P て))
(ADJI ほしい)
(PU 。))
(ID 264_textbook_kisonihongo))
The auxiliary い-adjective ほしい takes an IP-SMC complement with a controllee position that regularly has a non-subject controller. The IP-SMC is a (VB) (P て) or (VB0) (P て) sequence. See example (133).
The desiderative suffix たい inflects like an い-adjective, but as it is a bound suffix, it is assigned to the category AX (auxiliary).
There are auxiliary い-adjectives that follow the combining stems of verbs (viz. やすい、にくい, がたい). These are also assigne to the category AX.
There are い-adjectives that are formed with a verbal element but that end in the adjectival form づらい (e.g., しづらい, 分かりづらい, 見えづらい, etc.). These are treated as single words with the category ADJI.
Most copular expressions are formed by the combination [な-adjective/nominal expression + copula]. However, in cleft sentences the copula can combine with PP as well (see section 9.4.8 for details). Before going into detail about these types of expression, it is useful to consider the function of the copula, and to look at the forms and distribution of the copula. As also noted in the section 9.4.6, the copula functions semantically to associate a subject noun phrase with another noun in one of three basic ways: (i) identity (e.g., 遊び人の金さんは奉行の遠山景元だ); (ii) property ascription(e.g., 油揚げは狐色だ); (iii) specification (e.g., 勝つのは俺だ).
In this section the forms and distribution of the copula are described. だ and the various simple forms analogous to だ take the role of copula and are tagged AX. The analytic form である also functions like a copula. This form is analysed into
(... (AX で)
(VB2 ある))
The negated copula also is analyzed, and typically a toritate particle intervenes between components:
(... (AX で)
(P は)
(VB2 ない))
In addition to attaching to nominal expressions (NP-PRDs) and な-adjectives (ADJNs), だ and its analogous forms also attach to a clausal constituent (P の).
An affirmative non-past non-volitional simple copula だ is ommitted immediately before some MDs (e.g., らしい, みたい, かもしれない, にちがいありません, そう, べき, etc.), whereas the complex form である is possible. There exist volitional/presumptive copulas consisting of a copula stem (だろ, でしょ, であろ) immediately followed by (MD う).
(134)
怖いのは目立たない問題だろう。
( (IP-MAT (PP (NP (IP-REL (NP-SBJ *T*)
(ADJI 怖い))
(N の))
(P は))
(NP-SBJ *)
(NP-PRD (IP-REL (NP-SBJ *T*)
(VB 目立た)
(NEG ない))
(N 問題))
(AX だろ)
(MD う)
(PU 。))
(ID 31_news_KAHOKU_616))
Note that the copula (AX だろ) (MD う) is identical in pronunciation to (MD だろう). These volitional copulas do not co-occur before (MD だろう) and (MD だろう) never supplants them. (MD だろう) can appear after any predicate that doesn't involve a copula, and after AXD following any predicate (including after past tense copular expressions).
(135)
昨日は東北でも暖かかっただろう。
‘I suppose that it was also warm in the Tohoku region yesterday .’
( (IP-MAT (PP (NP (N 昨日))
(P は))
(NP-SBJ *)
(PP (NP (NPR 東北))
(P で)
(P も))
(ADJI 暖かかっ)
(AXD た)
(MD だろう)
(PU 。))
(ID 451_textbook_purple_intermediate))
An affirmative non-past non-volitional simple copula だ is omitted immediately before the conditional (P なら) (in contrast to the toritate particle なら), But note that there is a historically attested copula なる with a combining stem なら that is identical in pronunciation to (P なら). Also note that (P なら) can supplant the copula after nominal predicates, can appear after any predicate that doesn't involve a copula, and after AXD following any predicate (including after past tense copular expressions).
An affirmative non-past non-volitional simple copula だ is omitted immediately before some sentence-final particles (か, さ). Note that in embedded indirect questions, だ can be dropped, but the sequence だか is allowed.
(136)
「私は涙というものがどんなものかを知らなかった。
( (IP-MAT (-LRB- 「)
(PP (NP (PRO 私))
(P は))
(NP-SBJ *)
(PP (CP-QUE (IP-SUB (PP (NP (PP (NP (N 涙))
(P という))
(N もの))
(P が))
(NP-SBJ *が*)
(NP-PRD (WD どんな)
(N もの)))
(P か))
(P を))
(NP-OB1 *を*)
(VB 知ら)
(NEG なかっ)
(AXD た)
(PU 。))
(ID 78_aozora_Yuki-1-2000))
(137)
なぜだかは分かりません。
( (IP-MAT (NP-SBJ *speaker*)
(PP (CP-QUE (IP-SUB (NP-SBJ *pro*)
(ADVP (WADV なぜ))
(AX だ))
(P か))
(P は))
(VB 分かり)
(AX ませ)
(NEG ん)
(PU 。))
(ID 303_aozora_Hayashida-2015))
だ is optional before some sentence-final particles: (ね, よ). In all these cases of だ ommission, a null copula need not be supplemented.
On the other hand, there are sentence-final particles before which だ cannot be ommitted: ぞ, ぜ, わ, なぁ, etc. There is a modal element after which だ cannot be ommitted: hearsay そう. だ omission is also ruled out for many clause linkage particles: けれど, とも, から, し, が, etc.
Matrix predicate extension (P の) (AX だ) is sensitive to sentence-final particles in the same way that core copular predicates are: Long forms can appear, but the short form is regularly ommitted. Compare
(138)
太郎はまだ子供なのだ。
‘Taro is still a child.’
( (IP-MAT (PP (NP (NPR 太郎))
(P は))
(NP-SBJ *)
(ADVP (ADV まだ))
(NP-PRD (N 子供))
(AX な)
(FN の)
(AX だ)
(PU 。))
(ID 39_textbook_kisonihongo))
In the のだ construction the particle の is not annotated as if it were a head noun modified by a relative clause. This is due to various reasons: Exhaustive focus for が-marked subjects of stative predicates is not suppressed in the のだ construction, whereas it is suppressed in relative clauses. Scope effects discernable for various [formal noun + copula] constructions don't show up with のだ: c.f. 火曜日ならここはベルギーな筈だ vs. 火曜日ならここはベルギーなんだ. On the other hand, some unambiguously adnominal elements can appear in the のだ construction, whereas they must be adverbial without のだ : c.f. あいつ、あんなガメツイ人なんだ！ あいつ、あんな*(に)ガメツイ人だった！ Furthermore, there are cases where the scope does not include the entire sentence:
(139)
それはそうとしても、これがまだ彼の父親なのだろうか。
( (CP-QUE (IP-SUB (PP (IP-ADV (NP-SBJ *arb*)
(PP (NP (PRO それ))
(P は))
(NP-OB1 *)
(PP (ADVP (ADV そう))
(P と))
(VB し)
(P て))
(P も))
(SCON *)
(PU 、)
(PP (NP (PRO これ))
(P が))
(NP-SBJ *が*)
(ADVP (ADV まだ))
(NP-PRD (PP (NP (PRO 彼))
(P の))
(N 父親))
(AX な)
(FN の)
(AX だろ)
(MD う))
(P か)
(PU 。))
(ID 795_aozora_Harada-1960))
The policy is to err on the side of caution at least in this particular case, although a similar treatment might be advisable for various of the formal nouns that have modal meanings.
Copulas have a variety of phonological shapes, some of which are homophonous with particles. For example, the form で (in addition to supplying the stem for analytic copulas that get inflectional morphology from ある, いる, and ございます), can also appear on its own as a non-finite copula ending a non-final clause. For example,
(140)
＜ＳＬ銀河＞客車は４両編成で、定員１８０人を予定。
( (IP-MAT (PP (NP (PRN (-LRB- ＜)
(NP (NPR ＳＬ銀河))
(-RRB- ＞))
(N 客車))
(P は))
(NP-SBJ *)
(IP-ADV (NP-PRD (NUMCLP (NUM ４)
(CL 両))
(N 編成))
(AX で))
(CONJ *)
(PU 、)
(PP (NP (N 定員)
(NUMCLP (NUM １８０)
(CL 人)))
(P を))
(NP-OB1 *を*)
(VB 予定)
(PU 。))
(ID 30_news_KAHOKU_1917))
(141)
仙台は静かでとてもきれいでした。
‘Sendai was quiet and so beautiful.’
( (IP-MAT (PP (NP (NPR 仙台))
(P は))
(NP-SBJ *)
(IP-ADV (ADJN 静か)
(AX で))
(CONJ *)
(ADVP (ADV とても))
(ADJN きれい)
(AX でし)
(AXD た)
(PU 。))
(ID 9_misc_EXAMPLE))
Note how the short form of the copula (AX で) in (140) and in (141) can in principle be substituted with であって.
The phonological forms に and と appear in contexts where infinitive copulas are expected. The former frequently appears after ADJN, as in (142), but also with nominal predicates in small clause constructions (143). The latter also appears in the same condition (144). These に and と are analyzed as (AX に) and (AX と) in those contexts (see sections 34.1.1--34.1.3 for a detailed discussion).
(142)
子供は一生懸命に手拭を見ていた。
( (IP-MAT (PP (NP (N 子供))
(P は))
(NP-SBJ *)
(ADVP (ADJN 一生懸命)
(AX に))
(PP (NP (N 手拭))
(P を))
(NP-OB1 *を*)
(VB 見)
(P て)
(VB2 い)
(AXD た)
(PU 。))
(ID 261_aozora_Natsume-1908))
(143)
全学級の大騒ぎになった。
( (IP-MAT (NP-SBJ *pro*)
(IP-SMC (NP-PRD (PP (NP (QN 全学級))
(P の))
(N 大騒ぎ))
(AX に))
(VB なっ)
(AXD た)
(PU 。))
(ID 60_aozora_Dazai-1-1940))
(144)
そんなことを夫にこぼしても、馬耳東風で平然としている。
( (IP-MAT (PP (IP-ADV (NP-SBJ *speaker*)
(PP (NP (D そんな)
(N こと))
(P を))
(NP-OB1 *を*)
(PP (NP (N 夫))
(P に))
(VB こぼし)
(P て))
(P も))
(CND *)
(PU 、)
(NP-SBJ *pro*)
(IP-ADV (NP-PRD (N 馬耳東風))
(AX で))
(CONJ *)
(ADJN 平然)
(AX と)
(VB2 し)
(P て)
(VB2 いる)
(PU 。))
(ID 18_news_KAHOKU_10414))
A long form based on と is seen in some set phrases such as 「私ともあろう者が...」 and fossilized in the determiner とある (e.g., 渋谷のとある喫茶店でインタビューに応じてくださった).
Contractions of に and と with ある can form adnominal copulas:
(145)
あれは無意味なる沈鬱である。
( (IP-MAT (PP (NP (PRO あれ))
(P は))
(NP-SBJ *)
(NP-PRD (IP-REL (NP-SBJ *T*)
(ADJN 無意味)
(AX なる))
(N 沈鬱))
(AX で)
(VB2 ある)
(PU 。))
(ID 13_aozora_Terada-1929))
(146)
彼には確固たる考えがある。
( (IP-MAT (PP (NP (PRO 彼))
(P に)
(P は))
(NP-SBJ *に*)
(PP (NP (IP-REL (NP-SBJ *T*)
(ADJN 確固)
(AX たる))
(N 考え))
(P が))
(NP-OB1 *が*)
(VB ある)
(PU 。))
(ID 54_misc_EXAMPLE))
Given the function and distribution of the items described above, the assignment of a distinct part of speech AX seems more than justifiable.
A particularly difficult distinction to make is that between the copular use of the form の and its use as a particle. We present an operational definition below in order to make annotation and search unambiguous.
Among the instances of form の that modify nouns, there are those that should be identified as copulas. For such items, の is tagged as (AX の) as shown in (147) below, and the whole modifying phrase is treated as a relative clause.
(147)
「そこに小さなマッチ売りの少女がいる。
( (IP-MAT (-LRB- 「)
(PP (NP (PRO そこ))
(P に))
(PP (NP (PNL 小さな)
(IP-REL (NP-SBJ *T*)
(NP-PRD (N マッチ売り))
(AX の))
(N 少女))
(P が))
(NP-SBJ *が*)
(VB いる)
(PU 。))
(ID 225_aozora_Yuki-1-2000))
In contrast to these, there are instances of analogous strings where の is a post-positional particle, and in these の is tagged as (P の), as in (148). This practice is for indicating that the preceding noun is in some sense the argument of the following noun.
(148)
中世の都市の自治区には同業組合と商売の名前がありました
( (IP-MAT (PP (NP (PP (NP (PP (NP (N 中世))
(P の))
(N 都市))
(P の))
(N 自治区))
(P に)
(P は))
(PP (NP (PP (NP (CONJP (NP (N 同業組合))
(P と))
(NP (N 商売)))
(P の))
(N 名前))
(P が))
(NP-SBJ *が*)
(VB あり)
(AX まし)
(AXD た))
(ID 89_ted_talk_8))
The following is a test to identify the particle の: From an expression 「X1の NP2」, make a string of the form （この）NP2はX1{だ/だった}. If the string so formed is grammatical (allowing changes in propositional content), and if X1 either originally contains a core or role particle, or if by having a “functional element” added to X1 the propositional content of the initial expression is preserved unchanged, then let の in the initial expression be marked as P. For the purposes of this test, “functional element” is defined as either が, の, を, に, で, によって, から, まで or combinations of these.
This is a test to determine whether X1 can be an argument or an adjunct. If a “functional element” can be inserted, the NP to which this is applicable can be considered to be either an argument or an adjunct. Furthermore, this test comprises no more than one sufficient condition, so that its converse does not obtain. Accordingly, be aware that among strings that do not pass this test, there are items that nevertheless should be tagged as P (discussed below).
Among the strings that pass this test are such items as じゃがいもの皮むき; 物理の本; 太郎のかばん; 元旦の夕食; 学校の机; 押入れの中のふとん; 上北台までのモノレール; ドーナツとは別の食べ物.
As for strings that do not pass the test, other things being equal, they are tagged as AX and as such project a relative clause. Applicable strings include such items as 病気の太郎; これらの教科書; 立川においでのお客様; 学生の三人.
Quantifying expressions like that in 三人の学生 (three students) technically do not pass the test, and accordingly are subject to a copula analysis, but in order to distinguish these from property-denoting expressions such as 一尺の布 (cloth one foot in length), we tag の as P and indicate that the whole expression is quantifying by adding disambiguating information to the PP formed thereby: (PP;* (NP (NUMCLP (NUM 三) (CL 人))) (P の)). See section 29.2 for details.
Note that, among the reasons for a string failing the test, when the derived expression （この）NP2はX1{だ/だった} is ungrammatical in the first place, the items in question are treated as exceptions. Examples of these are ため (reason, cause); ため (beneficiary); せい (reason, cause); せい (wrongdoer) こと, よう. These are all examples of formal nouns. The annotation for these is described below.
For the expression X1のため (reason, cause), の is tagged as AX and the entire modifying phrase is analysed as an IP-EMB. In this usage, の can be replaced either by である or by をする to form X1{である/をする}ため without a change in propositional meaning. This test distinguishes the usage of X1のため for reason or cause from the usage for beneficiary, and its converse is probative. For example:
(149)
被災地ではまだ、保護者が大変な時期のため、子どもがうまく甘えられない面もある。
( (IP-MAT (PP (NP (N 被災地))
(P で)
(P は))
(NP-ADV (IP-EMB (ADVP (ADV まだ))
(PU 、)
(PP (NP (N 保護者))
(P が))
(NP-SBJ *が*)
(NP-PRD (IP-EMB (ADJN 大変)
(AX な))
(N 時期))
(AX の))
(N ため))
(PU 、)
(PP (NP (IP-EMB (PP (NP (N 子ども))
(P が))
(NP-SBJ *が*)
(ADVP (ADV うまく))
(VB 甘え)
(VB2 られ)
(NEG ない))
(N 面))
(P も))
(NP-SBJ *)
(VB ある)
(PU 。))
(ID 99_news_KAHOKU_55))
For the usage of X1のため to denote a beneficiary (where X1 is an NP), の is tagged as P, as in the example below:
(150)
この「毒殺未遂事件」の正体は、反政宗派一掃のための自作自演説もある。
( (IP-MAT (PP (NP (PP (NP (D この)
(-LRB- 「)
(N 毒殺未遂事件)
(-RRB- 」))
(P の))
(N 正体))
(P は))
(NP-SBJ *)
(PU 、)
(PP (NP (NML (PP (NP (PP (NP (N 反政宗派一掃))
(P の))
(N ため))
(P の))
(N 自作自演))
(N 説))
(P も))
(NP-OB1 *)
(VB ある)
(PU 。))
(ID 46_wikipedia_Datemasamune))
The same treatment is adopted for the expression X1のせい and the distinction in its usage. For the expression X1のこと, the practice is to tag の as P.
(151)
「政治家たちは子供たちのことにはまったく無関心です。
( (IP-MAT (-LRB- 「)
(PP (NP (N 政治家たち))
(P は))
(NP-SBJ *)
(PP (NP (PP (NP (N 子供たち))
(P の))
(N こと))
(P に)
(P は))
(ADVP (ADV まったく))
(ADJN 無関心)
(AX です)
(PU 。))
(ID 251_wikipedia_Audrey_Hepburn))
However, note that when the expression X1のこと appears in constructions such as 未提出者は放課後、居残りのこと, の is tagged as P and こと is tagged as MD. As for expressions like *太郎が該当者のこと（が重大だ）, although they are not expected to appear as examples of contemporary Japanese, when they are attested, the formal noun is treated as taking a modifying IP-EMB.
For expression of the form X1のよう, の is tagged as AX and the entire modifying phrase is treated as an IP-EMB. With regard to the item よう, it is a fact of the language that a preceding の can always be replaced with である.
(152)
猫はあたかも何事も起らなかったかのようにうそうそと橋の欄干を嗅いでいた。
( (IP-MAT (PP (NP (N 猫))
(P は))
(NP-SBJ *)
(IP-ADV (ADVP (ADV あたかも))
(NP-PRD (PP (CP-QUE (IP-SUB (PP (NP (N 何事))
(P も))
(NP-SBJ *)
(VB 起ら)
(NEG なかっ)
(AXD た))
(P か))
(P の))
(N よう))
(AX に))
(SCON *)
(PP (ADVP (ADV うそうそ))
(P と))
(PP (NP (PP (NP (N 橋))
(P の))
(N 欄干))
(P を))
(NP-OB1 *を*)
(VB 嗅い)
(P で)
(VB2 い)
(AXD た)
(PU 。))
(ID 95_aozora_Terada-1929))
(153)
これは非常に強烈な経験であたかも自分は存在しないかのように感じます
( (IP-MAT (NP-SBJ *speaker*)
(IP-ADV (PP (NP (PRO これ))
(P は))
(NP-SBJ *)
(NP-PRD (IP-REL (NP-SBJ *T*)
(ADVP (ADJN 非常)
(AX に))
(ADJN 強烈)
(AX な))
(N 経験))
(AX で))
(CONJ *)
(NP-OB1 *pro*)
(IP-SMC (ADVP (ADV あたかも))
(NP-PRD (PP (CP-QUE (IP-SUB (PP (NP (PRO 自分))
(P は))
(NP-SBJ *)
(VB 存在)
(VB0 し)
(NEG ない))
(P か))
(P の))
(N よう))
(AX に))
(VB 感じ)
(AX ます))
(ID 54_ted_talk_3))
The above proposal sets out how it is possible to choose between a PP analysis and an IP-EMB analysis with regard to formal nouns by applying the test of whether の can be replaced either by である or by をする. (If a flow chart could be provided, that would be even better. If tikz were made available this would be easy to make.)
While the policy of specifying the copular function of の (as well as other forms of the copula that are homophonous with particles) adds a burden to annotators, it is part of a program to simplify the description of the grammar according to function and distribution. It also makes a great difference towards clarifying the semantic function of constituents in context. A set of procedures that differentiates between nexus and event modification will no doubt isolate more instances of copulas. But this still needs to be developed.
な-adjectives (ADJNs, also known as 形容動詞) normally appear followed by some form of copula: だ, な, に, の, で, と, たる, なる, etc. (e.g., しずかだ/な, りこうだ/な). The resulting expression is a predicate, which, when appearing with a subject, projects an IP. The predicative nature of expressions such as these is unambiguous, and there are a variety of syntactic facts that support a clausal analysis. This form is set apart primarily by virtue of (i) having a special short form of the adnominal copula, な; (ii) resisting modification by nominal complements; and (iii) resisting direct marking by particles other than か and complementizers と, という, といった, etc. Forms which have traditionally been called たる-conjugation adjectives are ADJNs: 鬱々たる/と, 閑散たる/と, 決然たる/と, 騒然たる/と, 漠然たる/と, 呆然と, 満々たる/と, 黙々たる/と, 隆々たる/と. There are also forms which, while sharing the main attributes of ADJNs, can also take either adnominal な or adnominal の: 不思議な/の; 当たり前な/の, キレイ好きな/の, etc. These too are analyzed as ADJNs. (Note there are forms which can be modified and case marked in the same way as nouns but take a な copula when heading a nominal predicate, most notably the formal nouns よう, はず, and 風. These are treated as Ns. (see section 9.4.6). In general, all nominal expressions can take a な copula when appearing as core predicate in a のだ construction.) The form に immediately following an ADJN is automatically identified as a 断定助動詞 by the UniDic/LUW analysis used to produce trees for this corpus. Adding に, の, で, と, たる, and なる to the inventory of copula forms simplifies a description of particles homophonous to them and clarifies the semantics of sentences employing them. See section 9.4.1 for more discussion.
Predicates formed on ADJNs project clauses in the same way as verbs and い-adjectives:
(154)
いったん崩れたときに、パキンと折れるような状態にならないか心配だ。
( (IP-MAT (NP-SBJ *pro*)
(CP-QUE (IP-SUB (NP-SBJ *pro*)
(PP (NP (IP-EMB (ADVP (ADV いったん))
(VB 崩れ)
(AXD た))
(N とき))
(P に))
(PU 、)
(PP (NP (IP-REL (NP-SBJ *T*)
(NP-PRD (IP-EMB (PP (NP (N パキン))
(P と))
(VB 折れる))
(N よう))
(AX な))
(N 状態))
(P に))
(VB なら)
(NEG ない))
(P か))
(ADJN 心配)
(AX だ)
(PU 。))
(ID 34_news_KAHOKU_616))
(155)
太郎と花子では、どちらが歌が上手ですか。
‘Who sings better, Taro or Hanako?’
( (CP-QUE (IP-SUB (PP (NP (PP (NP (CONJP (NP (NPR 太郎))
(P と))
(NP (NPR 花子)))
(P で)
(P は))
(PU 、)
(WPRO どちら))
(P が))
(NP-SBJ *が*)
(PP (NP (N 歌))
(P が))
(NP-OB1 *が*)
(ADJN 上手)
(AX です))
(P か)
(PU 。))
(ID 335_textbook_kisonihongo))
(156)
家庭が不安定だと子どもの状態は良くならない。
( (IP-MAT (PP (IP-ADV (PP (NP (N 家庭))
(P が))
(NP-SBJ *が*)
(ADJN 不安定)
(AX だ))
(P と))
(CND *)
(PP (NP (PP (NP (N 子ども))
(P の))
(N 状態))
(P は))
(NP-SBJ *)
(ADJI 良く)
(VB2 なら)
(NEG ない)
(PU 。))
(ID 39_news_KAHOKU_616))
(141)
仙台は静かでとてもきれいでした。
‘Sendai was quiet and so beautiful.’
( (IP-MAT (PP (NP (NPR 仙台))
(P は))
(NP-SBJ *)
(IP-ADV (ADJN 静か)
(AX で))
(CONJ *)
(ADVP (ADV とても))
(ADJN きれい)
(AX でし)
(AXD た)
(PU 。))
(ID 9_misc_EXAMPLE))
(158)
またも不思議に思って彼を見つめた。
( (IP-MAT (NP-SBJ *speaker*)
(IP-ADV (PP (ADVP (ADV また))
(P も))
(IP-SMC (ADJN 不思議)
(AX に))
(VB 思っ)
(P て))
(CONJ *)
(PP (NP (PRO 彼))
(P を))
(NP-OB1 *を*)
(VB 見つめ)
(AXD た)
(PU 。))
(ID 93_aozora_Hayashida-2015))
(159)
『可哀そうな男だ。
( (IP-MAT (-LRB- 『)
(NP-SBJ *hearer*)
(NP-PRD (IP-REL (NP-SBJ *T*)
(ADJN 可哀そう)
(AX な))
(N 男))
(AX だ)
(PU 。))
(ID 156_aozora_Togawa-1937))
(160)
大衆が静かなことは日本人の特徴である。
( (IP-MAT (PP (NP (IP-EMB (PP (NP (N 大衆))
(P が))
(NP-SBJ *が*)
(ADJN 静か)
(AX な))
(N こと))
(P は))
(NP-SBJ *)
(NP-PRD (PP (NP (N 日本人))
(P の))
(N 特徴))
(AX で)
(VB2 ある)
(PU 。))
(ID 40_aozora_Hayashida-2015))
Note that, as with い-adjectives, when used adverbially the ren'yookee form (ADJN) (AX) can project an ADVP.
(161)
カメラ付き携帯電話の普及などで、一般の人々が写真や映像を撮影し提供する機会が飛躍的に増えた。
( (IP-MAT (PP (NP (PP (NP (N カメラ付き携帯電話))
(P の))
(N 普及))
(P など)
(P で))
(PU 、)
(PP (NP (IP-EMB (PP (NP (PP (NP (N 一般))
(P の))
(N 人々))
(P が))
(NP-SBJ *が*)
(PP (NP (CONJP (NP (N 写真))
(P や))
(NP (N 映像)))
(P を))
(NP-OB1 *を*)
(IP-ADV (VB 撮影)
(VB0 し))
(CONJ *)
(VB 提供)
(VB0 する))
(N 機会))
(P が))
(NP-SBJ *が*)
(ADVP (ADJN 飛躍的)
(AX に))
(VB 増え)
(AXD た)
(PU 。))
(ID 24_news_KAHOKU_39))
(162)
彼は公然と語っているのに、人々はこれに対して何も言わない。
( (IP-MAT (PP (IP-ADV (NP-OB1 *pro*)
(PP (NP (PRO 彼))
(P は))
(NP-SBJ *)
(ADVP (ADJN 公然)
(AX と))
(VB 語っ)
(P て)
(VB2 いる))
(P のに))
(SCON *)
(PU 、)
(PP (NP (N 人々))
(P は))
(NP-SBJ *)
(PP (NP (PRO これ))
(P に対して))
(NP-OB1 (WPRO 何)
(P も))
(VB 言わ)
(NEG ない)
(PU 。))
(ID 559_bible_new))
Some predicates formed on ADJNs such as 好きだ, 嫌いだ, and 上手だ take two が-marked NPs, only the first of which is analyzed as having a subject role. The second one is assigned the tag NP-OB1.
(163)
鈴木さんが一番英語が上手です。
‘Mr. Suzuki is the best at English among his peers .’
( (IP-MAT (PP (NP (NPR 鈴木さん))
(P が))
(NP-SBJ *が*)
(ADVP (ADV 一番))
(PP (NP (N 英語))
(P が))
(NP-OB1 *が*)
(ADJN 上手)
(AX です)
(PU 。))
(ID 493_textbook_purple_basic))
A careful distinction must be made between this kind of sentence and the so called double subject sentence. In the first kind, clear grammatical roles obtain between the が-marked NPs and the predicate. In the second kind, the leftmost NP-SBJ does not have a direct subject role See sections 9.2 and 18.2 for more discussion.
Both appear in the example below. 必要だ is a two-place predicate with 専用アプリのインストール as NP-OB1, while 簡単 （だ) heads a double subject construction.
(164)
スマホやタブレット端末は専用アプリのインストールが必要だが、操作は簡単。
( (IP-MAT (PP-TPC (NP (CONJP (NP (N スマホ))
(P や))
(NP (N タブレット端末)))
(P は))
(PP (IP-ADV (NP-SBJ *arb*)
(PP (NP (PP (NP (N 専用アプリ))
(P の))
(N インストール))
(P が))
(NP-OB1 *が*)
(ADJN 必要)
(AX だ))
(P が))
(CONJ *)
(PU 、)
(PP (NP (N 操作))
(P は))
(NP-SBJ *)
(ADJN 簡単)
(PU 。))
(ID 74_news_KAHOKU_97))
There is a bound suffix がち that attaches to the combining stems of verbs, but combines with the copula in the same pattern as is see with ADJNs.
There are words similar in form to な-adjectives, but are traditionally put under the category of rentaishi because they only ever occur in an adnominal position, although some of them can project clauses: 大きな, 小さな, 可笑しな. In isolation, they are tagged as PNL. When they take arguments or are modified (an extremely rare thing), they can be reanalyzed as ordinary structures headed by ADJN under PNLP, IP-REL, or IP-EMB, depending on the particular relation with the target of modification. See section 12 for details.
(165)
自分でも驚くほど大きな声が出ました。
( (IP-MAT (PP (NP (IP-REL (NP-SBJ *T*)
(PP (IP-ADV (NP-SBJ *speaker*)
(PP (NP (PRO 自分))
(P で)
(P も))
(VB 驚く))
(P ほど))
(ADJN 大きな)
(AX *))
(N 声))
(P が))
(NP-SBJ *が*)
(VB 出)
(AX まし)
(AXD た)
(PU 。))
(ID 104_news_KAHOKU_34))
(166)
そして、それよりもなお大きなわざを、お示しになるであろう。
( (IP-MAT (NP-SBJ *pro*)
(CONJ そして)
(PU 、)
(PP (NP (IP-REL (NP-SBJ *T*)
(PP (NP (PRO それ))
(P よりも))
(ADVP (ADV なお))
(ADJN 大きな)
(AX *))
(N わざ))
(P を))
(NP-OB1 *を*)
(PU 、)
(VB お示し)
(P に)
(VB2 なる)
(MD であろう)
(PU 。))
(ID 347_bible_new))
Nominal predicates are headed by N, NPR, Q, PRO, WPRO, etc. and are assigned the label NP-PRD. A nominal predicate combines with copula to either (i) denote a property of a subject NP, or (ii) predicate an identity relation with a subject NP, or (iii) specify a member of a group-denoting NP, or value of a property-denoting NP.
As with other predicates, when appearing with subjects, predicates formed on NP-PRD project clauses:
(167)
暗黒エネルギーの正体を探ることが現代宇宙論の最大のテーマです。
( (IP-MAT (PP (NP (IP-EMB (NP-SBJ *pro*)
(PP (NP (PP (NP (N 暗黒エネルギー))
(P の))
(N 正体))
(P を))
(NP-OB1 *を*)
(VB 探る))
(N こと))
(P が))
(NP-SBJ *が*)
(NP-PRD (PP (NP (N 現代宇宙論))
(P の))
(PP (NP (N 最大))
(P の))
(N テーマ))
(AX です)
(PU 。))
(ID 14_news_KAHOKU_11382))
(168)
－審査委員長一押しの提案は何ですか。
( (CP-QUE (IP-SUB (SYM －)
(PP (NP (IP-REL (NP-SBJ *T*)
(NP-PRD (N 審査委員長一押し))
(AX の))
(N 提案))
(P は))
(NP-SBJ *)
(NP-PRD (WPRO 何))
(AX です))
(P か)
(PU 。))
(ID 101_news_KAHOKU_52))
(169)
自分は虜だから、腰をかける訳に行かない。
( (IP-MAT (PP (NP (PRO 自分))
(P は))
(NP-SBJ *)
(PP (IP-ADV (NP-PRD (N 虜))
(AX だ))
(P から))
(SCON *)
(PU 、)
(PP (NP (N 腰))
(P を))
(NP-OB1 *を*)
(VB かける)
(MD 訳に行かない)
(PU 。))
(ID 296_aozora_Natsume-1908))
(170)
藤原氏は小保内氏の側近で、昨年３月に退任。
( (IP-MAT (PP (NP (NPR 藤原氏))
(P は))
(NP-SBJ *)
(IP-ADV (NP-PRD (PP (NP (NPR 小保内氏))
(P の))
(N 側近))
(AX で))
(CONJ *)
(PU 、)
(PP (NP (N 昨年)
(NUMCLP (NUM ３)
(CL 月)))
(P に))
(VB 退任)
(PU 。))
(ID 127_news_KAHOKU_28))
(171)
その間が、たっぷり一時間はあった様に思われます。
( (IP-MAT (NP-SBJ *speaker*)
(PP (NP (D その)
(N 間))
(P が))
(NP-OB1 *が*)
(PU 、)
(IP-SMC (NP-PRD (IP-EMB (ADVP (ADV たっぷり))
(PP;*SBJ* (NP (NUMCLP (NUM 一)
(CL 時間)))
(P は))
(VB あっ)
(AXD た))
(N 様))
(AX に))
(VB 思わ)
(VB2 れ)
(AX ます)
(PU 。))
(ID 249_aozora_Edogawa-1929))
(172)
国際学術コンペの審査委員長で、フロリダ大西洋大教授のフランク・シュニッドマンさんに聞いた。
( (IP-MAT (NP-SBJ *speaker*)
(NP-OB1 *pro*)
(PP (NP (IP-REL (NP-SBJ *T*)
(IP-ADV (NP-PRD (PP (NP (N 国際学術コンペ))
(P の))
(N 審査委員長))
(AX で))
(CONJ *)
(PU 、)
(NP-PRD (N フロリダ大西洋大教授))
(AX の))
(NPR フランク・シュニッドマンさん))
(P に))
(VB 聞い)
(AXD た)
(PU 。))
(ID 82_news_KAHOKU_52))
(173)
だとすると、彼が犯人である可能性は低い。
‘If that is the case, the possibility that he is the criminal is low.’
( (IP-MAT (CONJ だとすると)
(PU 、)
(PP (NP (IP-EMB (PP (NP (PRO 彼))
(P が))
(NP-SBJ *が*)
(NP-PRD (N 犯人))
(AX で)
(VB2 ある))
(N 可能性))
(P は))
(NP-SBJ *)
(ADJI 低い)
(PU 。))
(ID 176_textbook_kisonihongo))
(174)
「先生、この人が生れつき盲人なのは、だれが罪を犯したためですか。
( (CP-QUE (-LRB- 「)
(IP-SUB (NP-VOC (N 先生))
(PU 、)
(PP (NP (IP-EMB (PP (NP (D この)
(N 人))
(P が))
(NP-SBJ *が*)
(NP-ADV (N 生れつき))
(NP-PRD (N 盲人))
(AX な))
(N の))
(P は))
(NP-SBJ *)
(PU 、)
(NP-PRD (IP-EMB (PP (NP (WPRO だれ))
(P が))
(NP-SBJ *が*)
(PP (NP (N 罪))
(P を))
(NP-OB1 *を*)
(VB 犯し)
(AXD た))
(N ため))
(AX です))
(P か)
(PU 。))
(ID 748_bible_new))
Some Ns (including formal nouns はず, よう, and 風), can take な as a copula when used adnominally. In general, all nominal expressions can take a な copula when appearing as core predicate in a のだ construction, and when modifying various formal nouns (see discussion below).
The class of so-called の-adjectives (e.g., 赤, グレー, ピンク, 緑色, 紫, 大型, 最高, 多額, 当面, 別, 本当, etc.) is not recognised in the corpus. While usually appearing in property-ascriptions, most of these items exhibit the same properties as common nouns. Furthermore, they do not form an easily definable class, and accordingly are treated as Ns that head NP-PRDs.
There is also a class of formal nouns, each member of which has at least one grammatical function when immediately followed by a copula. Appearing at the ends of main clauses, they indicate various orientations of aspect, information structure, modality, evidentiality, etc. The basic members of this type are はず (plausibility); わけ (grounds, conclusion); もの (presupposition, habituality, inevitability); こと (obligation, fact); よう (simile, presumption). To these might be added others that appear in a variety of contexts: つもり, 気, 魂胆 (intention); ところ (pre-onset, perfective, hypothetical); ため (purpose, cause); せい, おかげ (cause); ほう (alternative); まま (current state) そう (contingent evidence); etc. Most of these do not allow が/の alternation in subject marking of a preceding adnominal clause (unless the copula following them is itself adnominal), but do allow exhaustive listing focus with が-marked subjects of property-denoting predicates. These facts point to the same analysis given to the のだ construction. But they are treated as nouns in the corpus.
Nominal predicates can be marked by certain toritate particles (viz., だけ, のみ, ばかり, ならでは, ぐらい, and なんか). In this scenario, the PP is immediately followed by disambiguating information: (NP-PRD *).
(175)
私が持っているのは本だけです。
‘This is the only book I have .’
( (IP-MAT (PP (NP (IP-REL (NP-OB1 *T*)
(PP (NP (PRO 私))
(P が))
(NP-SBJ *が*)
(VB 持っ)
(P て)
(VB2 いる))
(N の))
(P は))
(NP-SBJ *)
(PP-PRD (NP (N 本))
(P だけ))
(AX です)
(PU 。))
(ID 420_textbook_TANAKA))
In focused pseudocleft sentences, PPs of any type can appear in the same position as a nominal predicate. In such scenarios, no disambiguating information is needed. See section 30.5 for more discussion.
(176)
「わたしたちが信じるのは、もうあなたが話してくれたからではない。
( (IP-MAT (-LRB- 「)
(PP (NP (IP-EMB (NP-OB1 *pro*)
(PP (NP (PRO わたしたち))
(P が))
(NP-SBJ *が*)
(VB 信じる))
(N の))
(P は))
(NP-SBJ *)
(PU 、)
(ADVP (ADV もう))
(PP-PRD (IP-ADV (NP-OB1 *pro*)
(PP (NP (PRO あなた))
(P が))
(NP-SBJ *が*)
(VB 話し)
(P て)
(VB2 くれ)
(AXD た))
(P から))
(AX で)
(P は)
(NEG ない)
(PU 。))
(ID 283_bible_new))
Role particles followed by の occasionally modify nouns, and role particles followed by と occasionally modify predicates.
An argument can be made for extending a copular analysis to の and と in these contexts, given that the same forms with analogous functions appear, respectively, as adnominal and infinitive copulas following NP-PRD and ADJN. But the present practice is to tag の and と as particles in these contexts.
See section 9.4.3 for more discussion.
Aside from afterthoughts (倒置法 ---see section 33.4), there are basically three different elements that can follow core predicates in a main clause: (i) predicate extensions that follow the combining stem of an inflecting element; (ii) predicate extensions that follow a finite inflection; (iii) sentence-final particles (see section 8.7). The indirect passive morpheme and the causative morpheme technically belong to the first group, but are categorized as VB in the corpus, while the potential, spontaneous, and honorific uses of (ら)れる are tagged VB2. The のだ construction follows either core predicates or extensions in finite inflections, and so belongs to group 2, but it is treated separately. The rest of the members of groups 1 and 2 are categorized as AX, AXD, MD, PASS, or NEG. Note that forms of the copula (です, な, の, で, だっ, だろ, に, なる, と, たる, etc.) are assigned AX but actually comprise part of a core predicate (aside from their use in のだ constructions). See section 9.4.1 for details.
Auxiliary elements are subcategorised into AX, AXD, MD, PASS, and NEG. Some of these segments are complex, composed of sequences of words that have come to be used with fixed modal meanings. The forms that attach to combining stems of verbs and verb-inflecting auxiliaries are presented below:
Note that the form (ら)れる used as an indirect passive morpheme and the form (さ)せる used as a causative morpheme are assigned to the category VB.
Items of the category MD attach to finite inflections of predicates. Some of the more common items are introduced below:
There are many variations among the set of modal expressions. Only a partial list is provided here. Note also cases like そう and よう which have different meanings depending on the form to which they attach.
Some verbal heads of IP-REL take an auxiliary た, but ascribe a stative property to the modified N rather than (or aside from) participation in a past event. This only obtains in noun-modifying contexts, basically in property ascriptions. Almost all of the cases involve modification of Ns that have the semantic role of Patient with respect to the predicate heading the relative clause, the predicates being almost always either transitive, unaccusative, or passive. Some exceptions include verbs of dressing, encasing, enclosing, covering, where the modified N takes the role of Goal, but seems to undergo change of state. For references regarding this phenomenon, see 金水敏. 「連体修飾の 『 タ』 について」. 『日本語の名詞修飾表現』 くろしお出版, 1994; Kusumoto, Kiyomi. "The semantics of non-past ta in Japanese" MIT Working Papers in Linguistics 2001.
| |
|
|
| Section 8 |  |
Section 10 |