{ "paper_id": "Y98-1007", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T13:37:30.708163Z" }, "title": "THE TWO KINDS OF JAPANESE NEGATIVE Nai IN TERMS OF THEIR NPI LICENSING CONDITIONS", "authors": [ { "first": "Akira", "middle": [], "last": "Ikeya", "suffix": "", "affiliation": { "laboratory": "", "institution": "Toyo Gakuen University NTT Basic Research Labs", "location": {} }, "email": "" }, { "first": "Masahito", "middle": [], "last": "Kawamori", "suffix": "", "affiliation": { "laboratory": "", "institution": "Toyo Gakuen University NTT Basic Research Labs", "location": {} }, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "Traditionally, Japanese is said to have two kinds of negative word nai, meaning 'not' in English: the nai occurring after a verb or adjective and the nai with an independent status as a word. In contrast to traditional Japanese linguistics, most generative studies of Japanese have made no such distinction but rather focused their attention on such issues as the interaction between a scope of negation and a quantifier and the behavior of negative polarity items. In this paper we attempt to integrate all the proposals of each position by making use of both syntactic and semantic features. We also claim that the independent nai has a distinctly adjectival status with regard to its* semantic structure. 'John's brother's book' constitutes a single constituent NP, we can correctly predict that the sentences (43) and (44) are ungrammatical.", "pdf_parse": { "paper_id": "Y98-1007", "_pdf_hash": "", "abstract": [ { "text": "Traditionally, Japanese is said to have two kinds of negative word nai, meaning 'not' in English: the nai occurring after a verb or adjective and the nai with an independent status as a word. In contrast to traditional Japanese linguistics, most generative studies of Japanese have made no such distinction but rather focused their attention on such issues as the interaction between a scope of negation and a quantifier and the behavior of negative polarity items. In this paper we attempt to integrate all the proposals of each position by making use of both syntactic and semantic features. We also claim that the independent nai has a distinctly adjectival status with regard to its* semantic structure. 'John's brother's book' constitutes a single constituent NP, we can correctly predict that the sentences (43) and (44) are ungrammatical.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "The Japanese language has two kinds of negative word nai meaning 'not' in English: one is the nai occurring after a verb or adjective and the other is the nai which has an independent status as a word. The example of the latter is:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(1) Kokoni ringo wa nai here apple Nom not Here is no apple.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "which is the negation of:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(2) Kokoni ringo ga aru here apple Nom is Here is an apple.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "The former is exemplified by the sentence:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(3) watashi wa ringo o tabe-nai I Nom apple Obj eat not I do not eat an apple.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "where tabe-nai is the negation of the verb taberu (eat), and the following: where omoshiroku-nai is the negation of the adjective omoshiroi.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "While the traditional Japanese linguistics has generally treated the independent nai as an adjective, it regarded the nai which occurs after a verb as an auxiliary and the nai which occurs after an adjective as an adjective. Hattori (1950) raised a question as to this distinction and proposed to use a category-neutral term. He called the nai after a verb fuzoku-keishiki (dependent form) while he called the one after an adjective fuzoku-go (dependent word).", "cite_spans": [ { "start": 225, "end": 239, "text": "Hattori (1950)", "ref_id": "BIBREF20" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "In a strong contrast to traditional Japanese linguistics, almost all the studies in generative grammar have made no distinction between the nai occurring after a verb or an adjective with no independent status, and the nai having an independent status by giving a unified name NEG whose syntactic or semantic status is left unspecified. They have focused their attention on the interaction between a scope of negation and a quantifier or the issue of the behavior of negative polarity items.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "In this paper we attempt to integrate all the proposals , of each position by making use of both syntactic and semantic features. Moreover, it is our main claim that the independent nai has a distinctly adjectival status in terms of a semantic structure which we propose below.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "The paper consists of the following arguments. Firstly, the independent nai shows a semantic structure which is characteristic of ordinary adjectives. Secondly, it shares the same feature with the dependent nai in terms of the sensitivity to what is called negative polarity items, that is, NPI. Thus, it can be argued that the independent nai has an overlapping feature of being an adjective and at the same time of being sensitive to NPI.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "In English, as well as in Japanese, there is a group of so called \"degree adjectives\" whose interpretation is heavily dependent on linguistic or pragmatic contexts. One such contextual factor is called Thematic Dimension by Bartsch (1986 Bartsch ( /1987 . In addition to this dimension it is proposed in Ikeya (1991 Ikeya ( , 1996 that it is necessary to recognize two other such dimensions, which are termed the Comparative Contrastive dimension and the Degree dimension. Only after these three vectors are specified is it possible to determine the truth condition of a sentence which contains a degree adjective.", "cite_spans": [ { "start": 224, "end": 237, "text": "Bartsch (1986", "ref_id": "BIBREF18" }, { "start": 238, "end": 253, "text": "Bartsch ( /1987", "ref_id": null }, { "start": 304, "end": 315, "text": "Ikeya (1991", "ref_id": "BIBREF21" }, { "start": 316, "end": 330, "text": "Ikeya ( , 1996", "ref_id": "BIBREF22" } ], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "When we say he is good, the sentence has to be specified in terms of in what respect he is good, as compared or contrasted with whom he is good and to what degree he is good. For example, in he is very good at basketball for a short Japanese all these dimensions are expressed: at basketball is what we call the THEMATIC DIMENSION (TD), for a short Japanese is the so called COMPARATIVE CONTRASTIVE DIMENSION (CD) and very is our DEGREE DIMENSION (DD).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "In English TDs have the following varieties: As these examples show, in English TDs are expressed by such expressions as 'in terms of', 'as regards', or other prepositional phrases headed by 'of', 'at', 'about', 'off', 'in', etc. All these expressions give specification to adjectives in what respect John is good, fine or quick. It should be noticed that all . these expressions grammatically correspond to adverbials. It should also be mentioned that TD is not obligatory. In such a sentence as the business is very slow no TD is expressed.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "A degree adjective like tall implicitly encodes a comparison dimension like taller than X, with X being specified either by a linguistic or non-linguistic context. Take for example, the following sentences.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "Kare ni has the following feature structure: Language, Information and Computation (P'ACLIC12), 18-20 Feb., 1998, 69-81 (6) a. He, is tall. b. For a Japanese, he is tall.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "[ CAT I", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "In (6a), size \"tallness\" is always relative to some implicit measure such as the height of an average person and it is nonsense to talk about tallness except relative to such a comparison class. On the other hand, in (6b) a comparison class is explicitly encoded in the form of for a Japanese. This is the case of a linguistic specification of a comparison dimension, while (6a) is the case of contextual non-linguistic specification of a comparison dimension. We incorporate the idea of three dimensions of adjectives, into that of HPSG as outlined in Sag (1987, 1994) in order to explain a wider range of data. At the same time this is an attempt to expand the original architecture of the HPSG framework, where there is no mechanism to handle the three kinds of dimensions. According to an expanded version of HPSG, the adjective nai has the following feature structure. As shown below, the three dimensions are treated as predicate modifiers. By the feature [+neg] the lexical items from (a) to (e) except (b) can be placed together into one group. The first nai can be distinguished from omoshiroku-nai, yoma -nai, omowa-nai by the absence of a feature +compo, which means that a negative element occurs after a verb adjective stem forming a composite word. Omoshiroku-nai and yoma-nai can be differentiated by the features +adj. and +vb. We will not treat the feature +bridge in this paper. In this way all the ideas contained both in the traditional and generative linguistics are incorporated into this feature system. For example, Hattori's fuzoku-keishiki (dependent form) can be represented as [+vb, +compo, +neg] . It should be noted that a feature [+neg] is a semantic head feature, which plays a crucial role in the licensing of NPI, as we will note below.", "cite_spans": [ { "start": 553, "end": 569, "text": "Sag (1987, 1994)", "ref_id": null }, { "start": 1605, "end": 1624, "text": "[+vb, +compo, +neg]", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "Semantic Structure of Adjectives 2.1 Three Dimensions", "sec_num": "2" }, { "text": "Before we go into the problem of sensitivity of a Japanese NPI shika to a negative element we will state briefly how the simple sentence containing an negative nai can be described in terms of the semantic structure proposed in the foregoing section. 8 If this sentence has an emphatic stress on kare and is followed by a sentence in parentheses, it has the following semantic structure. If this semantic structure is translated into HPSG framework, our predicate corresponds to Head Daughter, subject to Complement Daughter, CD and DD to Adjunct Daughter, respectively. The sentence (7) can be generated by the following rule. (9) [", "cite_spans": [ { "start": 628, "end": 631, "text": "(9)", "ref_id": "BIBREF26" } ], "ref_spans": [], "eq_spans": [], "section": "\u2022 Negative Adjective nai and an NPI", "sec_num": "2.2" }, { "text": "We have shown that the semantic structure proposed above can be incorporated into the basic framework of HPSG. It does not mean, however, that we can dispense with the semantic structure: the implicit element of the contextual dimension cannot be incorporated into the HPSG architecture. Hereafter, therefore, both the basic framework of HPSG and the semantic structure of adjectives as outlined above will be employed as a basic theoretical tool for the description of Japanese NPIs. It is generally accepted in traditional Japanese linguistics in both generative and non-generative persuasions that shika functions as an NPI: it never occurs without a negative element like nai. It attaches itself to the preceding nominal element placing a focus on it. The exact meaning of (10), for example, is that 'there is only greed in him and no other spirit such as a noble spirit or a kind heart exists in him.' [4-neg] Since the feature [-I-NPI] which exists in shika is a head feature, its structure is shared with the phrase donyoku shika, which in turn is licensed by the negative feature in the head nai. Thus the grammaticality of (10) is guaranteed.", "cite_spans": [ { "start": 907, "end": 914, "text": "[4-neg]", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "SUBCAT <>] --+ H [LEX] , C, (A)*", "sec_num": null }, { "text": "On the other hand, (11) has the following feature structure. In (11) , in contrast to (10) , kare ni shika has [+NPI] , which is licensed by [+neg] in the head nai. This sentence means that there is greed only in him, not in any other person.", "cite_spans": [ { "start": 64, "end": 68, "text": "(11)", "ref_id": "BIBREF28" }, { "start": 86, "end": 90, "text": "(10)", "ref_id": "BIBREF27" }, { "start": 111, "end": 117, "text": "[+NPI]", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "SUBCAT <>] --+ H [LEX] , C, (A)*", "sec_num": null }, { "text": "Let us turn to the semantics of shika. Before we go on to define the meaning of NPI shika let us consider a simpler plain negative sentence like 12 Using a quantifier and letting a variable range over the set of (possible) attributes of a person, we can express the same condition as follows:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "(Vx E A)[(greed = x) E K)}\u2022", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "This says that for all the attributes that a person may have if an attribute is shown to be greed then this attribute does not belong to (the set of attributes of) kare. This is intuitively correct and satisfactory for our present purpose. One notices in this condition that the consequent -.(x E K) is clearly to be associated with the negative nai, and since K represents all the properties possessed by kare, it is natural to associate K with kare. The rest of the condition,", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "(Vx E A)[(greed = x) x .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "then, is attributable to donyoku-ga. Notice that the condition (... x ...) need not be negative, since ga is not an NPI. Now let us turn to our goal, the semantics of shika. The sentence (10) means, in a similar manner to the representation above, roughly the same as follows:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "(Vx E A)[(greed x) E K)]", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "This is to interpret the shika... nai sentence as saying that for all the attributes that a person may have if it is not greed then it does not belong to kare. This condition is logically equivalent, classically, to the condition that any attribute he has is identifiable as greed. We notice that the first condition is practically the same as the set-theoretical condition A-{greed}, so we can rephrase this condition to:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "(Vx)[x E (A -{greed}) E K)].", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "Looking at this formulation, it is not too difficult to see, just as in the case of ga above, that the meaning of shika can be seen as corresponding to the condition:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "(Vx)[x E (A -{greed}) (...x ...)].", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "Main part of its job is the condition (A -{greed}) which is to exclude, or select, depending on which side of the set you are on, those elements not identical to 'greed'. In other words, the meaning of shika is a sort of a selection, or exclusion, function that given an item and an appropriate domain, selects the set of elements that are not identical to, or sometimes incompatible with, the given item. Notice also that, unlike ga, shika requires the condition (... x ...) to be in negative since it is an NPI. But how do we decide the appropriate domain? This is where the contrastive dimension (CD) comes in. CD can be seen as defining a function that given an element, associates with it its possibly relevant 'comparable elements'. More formally, let us denote by D all the possible objects of a world, including such intensional objects as concepts and properties. Then for each object in this domain, Language, Information and Computation (P. ACLIC12), 18-20 Feb, 1998, 69-81 we cane think of a function that given the contextually defined contrastive dimension c, gives a set A C D of objects related, with respect to the comparability given by c, with the object in question. In other words we are thinking of a function G : D x CD -4 D such that:", "cite_spans": [ { "start": 1243, "end": 1245, "text": "-4", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "G(greed)(c) = A C D.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "From the universal domain D the function G selects the possible items which can be compared or contrasted with the particular item x, e.g. 'greed', in question and assigns them the status of the exponents of a comparative/contrastive dimension specifying at the same time that x exists. This means that G is a selection function which picks up possible candidates of contrastive or comparative dimension and at the same time an exclusion function which excludes all the items except a particular x by assigning them the status of exponents of a comparative or contrastive dimension. For example, there is a relevant domain where 'greed' is contrasted in sentence (10): 'ambition', 'noble spirit', 'good will', 'kind heart', and so forth.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "Now the meaning of the phrase donyoku shika in context c can be represented as: The function. F selects from the domain one particular item, that is, 'greed' by specifying its unique existence and excludes other items in the domain in question: 'ambition', 'noble spirit', 'good will', 'kind heart' and so on. Thus it plays the role of exclusion and selection function at the same time.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "[", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Semantics of shika", "sec_num": "2.4" }, { "text": "In the,, preceding two sections we have demonstrated the basic tools with which to describe the two kinds of nai in Japanese: the independent nai and dependent nai. The former shares the characteristics of ordinary adjectives, which the latter lacks. But on the other hand, both share an interesting character of being sensitive to NPI. In order to capture this fact, we have given two kinds of nai a common semantic feature [+neg] . In this section we will discuss the three ways in which a negative element is sensitive to the Japanese NPI shika. (13) [ We will show below how shika in (14) , where one of the constituent NPs in the sentence is within the scope of nai, is licensed within our framework. These sentences in (18) and (19) to (22) show that when the head daughter nai has a feature specification [-I-neg, -compo, -Fadj], it has a scope over not only its complement daughter and adjunct daughter but also over internal constituents NP or PP in a complement daughter.", "cite_spans": [ { "start": 425, "end": 431, "text": "[+neg]", "ref_id": null }, { "start": 549, "end": 553, "text": "(13)", "ref_id": "BIBREF30" }, { "start": 588, "end": 592, "text": "(14)", "ref_id": "BIBREF31" } ], "ref_spans": [], "eq_spans": [], "section": "Independent and Dependent nai", "sec_num": "3" }, { "text": "We have discussed how the independent nai acting as a head daughter is sensitive to its complement or adjunct daughter. In this section, we will deal with the problem of how the dependent nai in a head daughter behaves in relation to its complement and adjunct daughter.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Verb Stem -1-Dependent NAI", "sec_num": "3.2.1" }, { "text": "It should be noted that sira-nai and hanasa-nai have a subject noun and a sentence as complement daughters and those sentences judged ungrammatical are the ones in which internal constituents, that is, NP or PP is. within the scope of nai. The sentences (25), (26) are cases in point. On the other hand, the sentences (27), (28) are those where the two complements, the subject NP and the sentential complement daughters are not internal constituents.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Verb Stem -1-Dependent NAI", "sec_num": "3.2.1" }, { "text": "This means that a sentential NP or PP complement cah be a possible candidate of a target of the scope of the dependent nai but the internal NP or PP constituents cannot. In other words, the dependent nai which combines with a verb stem cannot refer to constituents within a sentential complement. [enpitu sanbon-shika motteitte ii to] iwa-na katta I-Top you-to-only pencil three-only take allowed Comp tell-not past I didn't tell you that you are allowed to take with you three pencils only.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Verb Stem -1-Dependent NAI", "sec_num": "3.2.1" }, { "text": "The sentence (30) has three complement daughters: PP watashiwa, anata ni and a VP enpitu o sanbon motte itte ii , with iwa-nai 'not say ' as a head daughter. Among the five sentences above, (30), (31), and (33) are judged grammatical while c and e are ungrammatical. The grammatical sentences are those in which the PP and VP are complement daughters while the ungrammatical ones are those in which the internal constituents of a VP are a possible candidate of the scope of the complex verb phrase consisting of a verb stem and nai. In a word, dependent nai cannot have a scope over internal constituents of a complement daughter.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Verb Stem -1-Dependent NAI", "sec_num": "3.2.1" }, { "text": "The following sentence is a case where a dependent nai occurs not with a verb stem but with an adjective stem.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "(35) *Uma-ga biiru-shika nomu koto-ga omoshiroku-nai.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "horse-Nom beer-only drink fact-Nom interesting-not The fact that a horse drinks only beer is not interesting.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "(36) *Uma-shika biiru-o nomu koto-ga omoshiroku-nai. horse-only beer-Obj drink fact-Nom interesting-not", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "The fact that only a horse drinks beer is not interesting.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "(37) Uma-ga biiru-o nomu koto-shika omoshiroku-nai. horse-Nom beer-Obj drink fact-only interesting-not Only the fact that a horse drinks beer is interesting.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "It should be noticed that the basic structure of the sentences above consists of a complex predicate having a dependent nai as a head and a sentential complement uma ga biiru o nomu (a horse drinks beer) nominalized by koto. If the dependent nai refers to an NP constituent within an NP nominalized by koto, that is, when an internal NP is within the scope of a dependent nai , the sentence is judged as ungrammatical: the sentences (35) and (36) are cases in point. On the other hand, when the nai refers only to the sentential complement, the sentence is grammatical, which is exemplified by (37). How the NPI is licensed is shown below. We can summarize this section as follows: If a head daughter has a feature specification neg, adj, -compo], that is, if a head is an independent nai , the negative element has a scope over not only its complement or adjunct daughter, but also internal NP or PP constituents. On the other hand, if a head daughter has a feature specification [+neg, +vb,+adj, +compo], such a predicate has a limited scope: it has a scope over only its complement, not any internal NP or PP constituents. The difference can be attributed to the independent or dependent property of a negative element nai: in the case of an independent adjective nai the only element preceding it is the one with less semantic content such as a nominalizer koto but on the other hand, the dependent nai is preceded by a verb or adjective stem with a substantial semantic content. This generalization accords well with native speaker's intuition.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Adjective Stem + Dependent Nai", "sec_num": "3.2.2" }, { "text": "In this paper we have proposed the following.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Theoretical Implications and Conclusions", "sec_num": "4" }, { "text": "(1) Correspondence Between Syntactic and Semantic Structures of an Independent Nai In dealing with a semantic structure of an adjective we have stipulated three components: a predicate, argument, and predicate modifier. In the framework of HPSG a predicate corresponds to a head daughter, an argument to a complement daughter, and a predicate modifier to an adjunct . In terms of HPSG framework a head daughter nai has a scope over a complement and adjunct daughter. Equivalently, we can restate that a predicate have a scope over an argument and a predicate modifier in terms of a predicate-argument structure.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Theoretical Implications and Conclusions", "sec_num": "4" }, { "text": "(3) The Scope Principle 'In the scope of is defined over a head-complement with or without an adjunct daughter or a predicate argument structure in the case of independent nai, both of which are in an equivalent relation. The scope of a head daughter consists of its complement and or adjunct(s), and possibly their internal constituent(s).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Theoretical Implications and Conclusions", "sec_num": "4" }, { "text": "The semantic feature [-I-NPI} in a shika phrase is absorbed by [-Fneg] feature in a head daughter whenever [+NPI] is in the scope of [+neg] in the head daughter. This mechanism ensures that there must be only one shika present in the scope of a head daughter with [ -1-neg]. The first assumption implicitly states that there is a one-to-one mapping relation between a syntactic structure and a semantic counterpart, a basic claim made by traditional Montagovian frameworks. By the second and fifth principle it is easy to make a general statement why sentences with two shika's like (38) and (39) are grammatical, while the sentence (40) is judged as ungrammatical. In an embedded clause asu shika ame ga huranai of (38), shika is attached to an adjunct asu 'tomorrow' and the feature [-I-NPI] in asu shika 'only tomorrow' is absorbed by a negative feature H-neg] in the head daughter huranai 'rain not' because it is in the scope of a negative predicate in a minimal domain, that is, in a clause. The [+NPI] in the . phrase watashi shika is absorbed by the predicate iwanakatta 'said not' since it is in the scope of a negative predicate iwanakatta 'said not' in the next higher domain. In (39) the negative predicate hanasarete inai 'is spoken not' has an adjunct daughter, sono chihoo de shika, to which shika is attached. In a matrix clause a complement daughter [[sono chihoo de shika hanasarete inaalhoogen shika] the feature [+NPI] in hoogen shika is absorbed by a head daughter rikai dekinai so that the sentence as a whole is judged as grammatical. In (38) NPI] in the phrase asu shika is absorbed by a negative particle huranai 'rain not' so that there is no counterpart by which [+ NPI] in the phrase watashi shika is absorbed. Hence ungrammaticality of (38) results.", "cite_spans": [ { "start": 1432, "end": 1438, "text": "[+NPI]", "ref_id": null }, { "start": 1561, "end": 1565, "text": "(38)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "PP (4) NPI absorption", "sec_num": null }, { "text": "The third principle plays a most important role in this framework in that it attempts to define the licensing condition of an NPI without having recourse to the notion of configurational definition such as a c-command relation. If we follow the configurational principle, we are going to specify all the constituents in a clause as being in a c-command domain.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "PP (4) NPI absorption", "sec_num": null } ], "back_matter": [], "bib_entries": { "BIBREF1": { "ref_id": "b1", "title": "children-to English only taught fact-Obj I-Top know-not I don't know that John", "authors": [ { "first": "", "middle": [], "last": "John-Nom", "suffix": "" } ], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-Nom children-to English only taught fact-Obj I-Top know-not I don't know that John. taught only English to children.", "links": null }, "BIBREF3": { "ref_id": "b3", "title": "children-to English only taught fact-Obj he-Top talk-not John doesn't talk that he taught only English to children", "authors": [ { "first": "", "middle": [], "last": "John-Nom", "suffix": "" } ], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-Nom children-to English only taught fact-Obj he-Top talk-not John doesn't talk that he taught only English to children.", "links": null }, "BIBREF4": { "ref_id": "b4", "title": "In terms of a negative element both sire-nai, hanasa-nai consist of two components: a verb stem sira, hanasa and nai. We will see below how shika is distributed over the constituents of a sentence", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "In terms of a negative element both sire-nai, hanasa-nai consist of two components: a verb stem sira, hanasa and nai. We will see below how shika is distributed over the constituents of a sentence.", "links": null }, "BIBREF5": { "ref_id": "b5", "title": "John-only children-to English-Obj taught fact-Obj I-Top know-not I don't know that only John taught English to children", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-only children-to English-Obj taught fact-Obj I-Top know-not I don't know that only John taught English to children.", "links": null }, "BIBREF6": { "ref_id": "b6", "title": "Obj taught fact-Obj I-Top know-not I don't know that John taught English only to children", "authors": [ { "first": "", "middle": [], "last": "John-Only Children-To Only", "suffix": "" }, { "first": "", "middle": [], "last": "English", "suffix": "" } ], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-only children-to only English-Obj taught fact-Obj I-Top know-not I don't know that John taught English only to children. (27) [John-ga kodomotati-ni eigo osieta koto-o] lwatasi-shika] sir-anai.", "links": null }, "BIBREF7": { "ref_id": "b7", "title": "John-only children-to English taught fact-Obj I-only know-not No one but me knows that John taught English to children", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-only children-to English taught fact-Obj I-only know-not No one but me knows that John taught English to children. (28) [John-ga kodomotati-ni eigo osieta koto-shika] [watasi-wa] sir-anai.", "links": null }, "BIBREF8": { "ref_id": "b8", "title": "John-only children-to English taught fact-only I-Top know-not I know only the fact", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-only children-to English taught fact-only I-Top know-not I know only the fact that John taught English to children.", "links": null }, "BIBREF9": { "ref_id": "b9", "title": "Unless there is some ad-hoc restriction, all the constituents in a clause are in the c-command domain of the negative predicate yomanai 'read not'. For example, of the following four sentences sentences, 43 and 44 must be filtered out as ungrammatical by some means or other", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Unless there is some ad-hoc restriction, all the constituents in a clause are in the c-command domain of the negative predicate yomanai 'read not'. For example, of the following four sentences sentences, 43 and 44 must be filtered out as ungrammatical by some means or other.", "links": null }, "BIBREF10": { "ref_id": "b10", "title": "John-only he-Gen brother-Gen book-Obj read-not Only John reads his brother's book", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John-only he-Gen brother-Gen book-Obj read-not Only John reads his brother's book.", "links": null }, "BIBREF12": { "ref_id": "b12", "title": "John he-Gen-only brother-Gen book-Obj read-not", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John he-Gen-only brother-Gen book-Obj read-not", "links": null }, "BIBREF14": { "ref_id": "b14", "title": "John he-Gen brother-Gen-only book-Obj read-not", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John he-Gen brother-Gen-only book-Obj read-not", "links": null }, "BIBREF16": { "ref_id": "b16", "title": "John he-Gen brother-Gen book-Obj-only read-not John reads only his brother's book", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "John he-Gen brother-Gen book-Obj-only read-not John reads only his brother's book.", "links": null }, "BIBREF17": { "ref_id": "b17", "title": "On the other hand, in our framework 'yomanai' belongs to a class of verbs which cannot have a scope over internal constituents of a complement daughter", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "On the other hand, in our framework 'yomanai' belongs to a class of verbs which cannot have a scope over internal constituents of a complement daughter, and because John no otooto no hon References", "links": null }, "BIBREF18": { "ref_id": "b18", "title": "The construction of properties under perspective", "authors": [ { "first": "Renata", "middle": [], "last": "Bartsch", "suffix": "" } ], "year": 1986, "venue": "Journal of Semantics", "volume": "5", "issue": "", "pages": "293--320", "other_ids": {}, "num": null, "urls": [], "raw_text": "Bartsch, Renata (1986/ 87) \"The construction of properties under perspective,\" Journal of Se- mantics 5, 293-320.", "links": null }, "BIBREF19": { "ref_id": "b19", "title": "Language in a Psychological Setting", "authors": [ { "first": "Noam", "middle": [], "last": "Chomsky", "suffix": "" } ], "year": 1986, "venue": "Sophia Linguistica", "volume": "22", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Chomsky, Noam. (1986) Language in a Psychological Setting. Sophia Linguistica 22, Monograph. Tokyo: Sophia University.", "links": null }, "BIBREF20": { "ref_id": "b20", "title": "Fuzokugo to fuzoku-keishiki", "authors": [ { "first": "Shiro", "middle": [], "last": "Hattori", "suffix": "" } ], "year": 1950, "venue": "Gengo Kenkyuu", "volume": "15", "issue": "", "pages": "1--26", "other_ids": {}, "num": null, "urls": [], "raw_text": "Hattori, Shiro (1950) \"Fuzokugo to fuzoku-keishiki, \" Gengo Kenkyuu 15, 1-26.", "links": null }, "BIBREF21": { "ref_id": "b21", "title": "A contextual approach to Japanese Adjectives", "authors": [ { "first": "Akira", "middle": [], "last": "Ikeya", "suffix": "" } ], "year": 1991, "venue": "The Sixth Japanese-Korean Joint Conference on Formal Linguistics", "volume": "", "issue": "", "pages": "64--90", "other_ids": {}, "num": null, "urls": [], "raw_text": "Ikeya, Akira (1991) \" A contextual approach to Japanese Adjectives, \" The Sixth Japanese- Korean Joint Conference on Formal Linguistics. ed. by Akira Ikeya, 64-90.", "links": null }, "BIBREF22": { "ref_id": "b22", "title": "Tough constructions in HPSG framework", "authors": [ { "first": "Akira", "middle": [], "last": "Ikeya", "suffix": "" } ], "year": 1996, "venue": "Meaning and Discourse -A Festschrift for Proffessor Eva Hajicove", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Ikeya, Akira (1996) \"Tough constructions in HPSG framework, \"In Meaning and Discourse -A Festschrift for Proffessor Eva Hajicove, ed. by Barbara Partee and Petr Sgall, Amsterdam , John Benjamins.", "links": null }, "BIBREF23": { "ref_id": "b23", "title": "Sika-nai constrution and a theory of binding", "authors": [ { "first": "Yasuhiko", "middle": [], "last": "Kato", "suffix": "" } ], "year": 1979, "venue": "Sophia Linguistica", "volume": "5", "issue": "", "pages": "92--104", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kato, Yasuhiko (1979) \"Sika-nai constrution and a theory of binding,\" Sophia Linguistica 5, 92-104.", "links": null }, "BIBREF24": { "ref_id": "b24", "title": "Negative Sentences in Japanese", "authors": [ { "first": "Yasuhiko", "middle": [], "last": "Kato", "suffix": "" } ], "year": 1985, "venue": "Sophia Linguistica", "volume": "19", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kato, Yasuhiko (1985) Negative Sentences in Japanese. Sophia Linguistica 19, Monograph. Tokyo: Sophia University.", "links": null }, "BIBREF25": { "ref_id": "b25", "title": "Negative polarity and movement\" Formal approaches to Japanese linguistics 1, MIT working papers in linguistics 24", "authors": [ { "first": "Yasuhiko", "middle": [], "last": "Kato", "suffix": "" } ], "year": 1994, "venue": "", "volume": "", "issue": "", "pages": "101--120", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kato, Yasuhiko (1994) \"Negative polarity and movement\" Formal approaches to Japanese lin- guistics 1, MIT working papers in linguistics 24, 101-120.", "links": null }, "BIBREF26": { "ref_id": "b26", "title": "Remarks on `-wa and -ga' from the standpoint of inference", "authors": [ { "first": "Masahito", "middle": [], "last": "Kawamori", "suffix": "" } ], "year": 1991, "venue": "Gengo", "volume": "11", "issue": "", "pages": "54--69", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kawamori, Masahito. (1991) \"Remarks on `-wa and -ga' from the standpoint of inference\" , Gengo 11, 54-69 .", "links": null }, "BIBREF27": { "ref_id": "b27", "title": "Exhaustive-listing -ga Accomodated", "authors": [ { "first": "Masahito", "middle": [], "last": "Kawamori", "suffix": "" } ], "year": 1993, "venue": "4th Interanational Conference on Pragmatics", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kawamori, Masahito. (1993) \"Exhaustive-listing -ga Accomodated\", 4th Interanational Confer- ence on Pragmatics.", "links": null }, "BIBREF28": { "ref_id": "b28", "title": "Hilbertian Description and Discourse Semantics", "authors": [ { "first": "Masahito", "middle": [], "last": "Kawamori", "suffix": "" } ], "year": 1996, "venue": "Sophia Linguistica", "volume": "39", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kawamori, Masahito. (1996) Hilbertian Description and Discourse Semantics. Sophia Linguistica 39, Monograph. Tokyo: Sophia University.", "links": null }, "BIBREF29": { "ref_id": "b29", "title": "Sika nai koobun no koozo\" (The structure of Sika nai construction) in Ohye Saburoo Sensei Tuitoo", "authors": [ { "first": "Keiji", "middle": [], "last": "Konomi", "suffix": "" } ], "year": 1989, "venue": "", "volume": "", "issue": "", "pages": "369--385", "other_ids": {}, "num": null, "urls": [], "raw_text": "Konomi, Keiji (1989) \" Sika nai koobun no koozo\" (The structure of Sika nai construction) in Ohye Saburoo Sensei Tuitoo Ronbunsyuu Hensyuu iinkai, ed. Eigogaku no siten, Kyuushu Daigaku syuppankai, 369-385.", "links": null }, "BIBREF30": { "ref_id": "b30", "title": "Negative polarity items in Japanese and English", "authors": [ { "first": "Susumu", "middle": [], "last": "Kuno", "suffix": "" } ], "year": 1994, "venue": "Proceedings of 1994 Nanzan Symposium", "volume": "", "issue": "", "pages": "225--264", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kuno, Susumu (1994) \"Negative polarity items in Japanese and English, Proceedings of 1994 Nanzan Symposium. 225-264.", "links": null }, "BIBREF31": { "ref_id": "b31", "title": "The sika nai construction and predicate restructuring", "authors": [ { "first": "Masatake", "middle": [], "last": "Muraki", "suffix": "" } ], "year": 1978, "venue": "Problems in Japanese Syntax and semantics", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Muraki, Masatake (1978) \"The sika nai construction and predicate restructuring,\" in John Hinds and Irwin Howard, eds., Problems in Japanese Syntax and semantics, Kaitakusha, Tokyo.", "links": null }, "BIBREF32": { "ref_id": "b32", "title": "Information-based syntax and semantics", "authors": [ { "first": "Carl", "middle": [], "last": "Pollard", "suffix": "" }, { "first": "", "middle": [], "last": "Ivan", "suffix": "" }, { "first": "; Csli", "middle": [], "last": "Sag", "suffix": "" }, { "first": "", "middle": [], "last": "Standford", "suffix": "" } ], "year": 1987, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Pollard, Carl and Ivan, Sag (1987 ) Information-based syntax and semantics, CSLI Standford.", "links": null }, "BIBREF33": { "ref_id": "b33", "title": "Head-driven phrase structure", "authors": [ { "first": "Carl", "middle": [], "last": "Pollard", "suffix": "" }, { "first": "", "middle": [], "last": "Ivan", "suffix": "" }, { "first": ";", "middle": [], "last": "Sag", "suffix": "" }, { "first": "Stanford", "middle": [], "last": "Csli", "suffix": "" } ], "year": 1994, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Pollard, Carl and Ivan, Sag (1994) Head-driven phrase structure. CSLI , Stanford.", "links": null }, "BIBREF34": { "ref_id": "b34", "title": "The projection of phrase structure and argument structure in Japanese, Studies on the universality of constraint-based phrase structure grammars", "authors": [ { "first": "Peter", "middle": [], "last": "Sells", "suffix": "" } ], "year": 1996, "venue": "", "volume": "", "issue": "", "pages": "39--90", "other_ids": {}, "num": null, "urls": [], "raw_text": "Sells, Peter (1996) \" The projection of phrase structure and argument structure in Japanese, Studies on the universality of constraint-based phrase structure grammars. 39-90.", "links": null } }, "ref_entries": { "FIGREF0": { "text": "Kono hon wa omoshiroku-nai this book Nom interesting not This book is not interesting.", "num": null, "type_str": "figure", "uris": null }, "FIGREF1": { "text": "donyoku shika], = Fgreed(G(greed)(c)) = Fgreed( A ) = AN[(Vx)[x E (A -{greed}) --+ N(x)]] This\u2022condition will allow a compositional interpretation of shika nai construction without positing such a global operation as shika ... nai lowering.", "num": null, "type_str": "figure", "uris": null }, "TABREF6": { "content": "
[[ \u2022 [ CAT SYNSEM CONTENTHEAD SUBCAT {PP[1]\u2022II NP[1]} ] RELN F([1.]) 1 NPI + j
Kare ni in (10) above has the following feature structure:
. CAT{PP}
[ PRED. MOD:kare ni (in him)
CONTENT CONTEXT DIM: TD
THEM REL:LOC
As shown above, PP donyoku-shikaSUBCAT{NP[SBJ]}
NP-FNPI]
kare-nidonyokushikanai
", "html": null, "type_str": "table", "num": null, "text": "2.3 The Feature Structure of ShikaShika has the following feature structure: as a head daughter it has a SUBCAT value PP or NP and has [-I-NPI1 as a semantic feature, which in effect ensures that it combines with a lexical item having [I-neg] as a semantic head feature. Shika is characterized as follows: kare ni functions as a predicate modifier modifying a predicate nai and its contextual dimension is a TD. Its thematic relation denotes an abstract location.The sentence (10) has the following feature structure." }, "TABREF9": { "content": "
(17) [ Merii wa kodomotati ni eigoosieta koto]shika nai
Mary only childrento Merii-wa kodomotati-ni eigo-o osieta koto shika naiS[SUBCAT{0}]
NP[+NPI]nai [SUBCAT{NP(SUBJ)}][A-neg]
NPshika 4-NPI]
Merii wa kodomotati ni eigo o osieta koto
Merii wa] [kodomotati ni] [eigooosieta koto]]ganai
MaryTop childrento English Obj taught experience Nom not
Mary has no experience of having taught English to children.
As the bracketing shows, there are four possibilities of occurrence of shika in the sentence above.
The following are examples:
(14) [ Merii shika kodomotati ni eigooosieta koto]ganai
Mary only childrento English Obj taught experience Nom not
Only Mary has the experience of having taught English to children.
(15) [Merii wa kodomotati ni shika eigooosieta koto]ga nai
Mary Top childrento only English Obj taught experience Nom not
Mary has the experience of having taught English only to children.
(16) [ Merii wa kodomotati ni eigoshika osieta koto]ga nai
Mary only childrento English only taught experience Nom not
Mary has the experience of having taught only English to children.
", "html": null, "type_str": "table", "num": null, "text": "English taught experience only not Mary has only the experience of having taught English to children.There are several points to be noted. One is that nai in the sentences above is a case of independent nai. As an independent nai it is a one-place adjective taking a nominalized sentence [Meth wa kodomotati ni eigo o osieta koto], meaning ' the fact that Mary has taught English to children.' This is indicated by koto in Japanese. Furthermore, when nai acts as a predicate the feature [+nai ] has a scope over every internal constituent NP or PP constituent within the nominalized sentence as well as the nominalized sentence itself. This is contrastive to the case where nai is a dependent element of a verb or an adjective.(17)" }, "TABREF10": { "content": "
(19) [Merii-ni shika kodomotati ni eigooosieta koto]ganai
Mary-(20) [Merii-ni wa kodomotati ni shika eigooosieta koto]ganai
Mary-Dat Top childrento only English Obj taught experience Nom not
For Mary there is an experience of having taught English only to children.
(21) [Merii-ni wa kodomotati ni eigoshika osieta koto]ganai
Mary-Dat Top childrento English only taught experience Nom not
For Mary there is an experience of having taught only English to children.
(22) [Merii-ni wa kodomotati ni eigoosieta koto]shika nai
Mary-Dat Top childrento English taught experience only not
For Mary there is only an experience of having taught English to children.
S[SUBCAT{O}]
NP[+NM]S[NP(SUBJ)]
NPSUBJ shika[A-NP]PPnai 4-ned
NP
VP[SUBCAT{NP}]Nominlizer
Meriikodomo-tati-ni eigo-o osietakotoga
(18) [1 Merii-ni-wa] [[kodomotati-ni] [eigo-o}osieta koto]]-ganai.
Mary-Dat-Top children-toEnglish-Obj taught experience-Nom no
", "html": null, "type_str": "table", "num": null, "text": "As shown in the diagram above, the [+neg] feature in the nai licenses one of the constituents within the complements of the head nai. In other words, as long as the head daughter has [+neg, -compo,+ adj] .features, it has a scope over any NP or PP constituent within the complement daughter. This is the first tentative generalization.In Japanese there is a slightly different sentence, where a PP Merii ni wa is employed instead of a subject Merii wa as shown below. For Mary there is no experience for having taught English to children.We treat this PP as an adjunct phrase denoting an abstract location: literally it means that \"in Mary there is no experience of having taught English to children.\" Kodomotati ni eigo o osieta koto as a whole denotes a verb phrase, not a sentence nominalized by koto.The distribution of shika is shown in the following sentences. Dat only children to English Obj taught experience Nom not Only for Mary there is an experience of having taught English to children." }, "TABREF13": { "content": "
(39) watashi-wa [[sono chihoo-de-shika hanas-arete i-nailhoogen-shika rikai dekinai.
I-Topthat area-in-onlyspeak-Passive be-not dialect-onlyunderstand cannot
I can undersand only a dialect which is spoken only in that area.
(40) *watashi-shika anata-ni [asu-shika ame-ga hur-anai]-to itta
I-onlyyou-Dat [tomorrow-only rain-Nom fall-not] -Comp say Past
", "html": null, "type_str": "table", "num": null, "text": "(38) watashi-shika anata-ni [asu-shika ame-ga hur-anai] -to iwa-na katta I-only you-Dat. [tomorrow-only rain-Nom fall-not] -Comp say-not Past Only I and no other person said to you that it will rain only tomorrow." } } } }