Argument Realization and Case in Japanese
Abstract and Keywords
Japanese is a language where the grammatical status of arguments and adjuncts is marked exclusively by postnominal case markers, and various argument realization patterns can be assessed by their case marking. Since Japanese is categorized as a language of the nominative-accusative type typologically, the unmarked case-marking frame obtained for transitive predicates of the non-stative (or eventive) type is ‘nominative-accusative’. Nevertheless, transitive predicates falling into the stative class often have other case-marking alignments, such as ‘nominative-nominative’ and ‘dative-nominative’. Consequently, Japanese provides much more varying argument realization patterns than those expected from its typological character as a nominative-accusative language.
In point of fact, argument marking can actually be much more elastic and variable, the variations being motivated by several linguistic factors. Arguments often have the option of receiving either syntactic or semantic case, with no difference in the logical or cognitive meaning (as in plural agent and source agent alternations) or depending on the meanings their predicate carry (as in locative alternation). The type of case marking that is not normally available in main clauses can sometimes be obtained in embedded contexts (i.e., in exceptional case marking and small-clause constructions). In complex predicates, including causative and indirect passive predicates, arguments are case-marked differently from their base clauses by virtue of suffixation, and their case patterns follow the mono-clausal case array, despite the fact that they have multi-clausal structures.
Various case marking options are also made available for arguments by grammatical operations. Some processes instantiate a change on the grammatical relations and case marking of arguments with no affixation or embedding. Japanese has the grammatical process of subjectivization, creating extra (non-thematic) major subjects, many of which are identified as instances of ‘possessor raising’ (or argument ascension). There is another type of grammatical process, which reduces the number of arguments by virtue of incorporating a noun into the predicate, as found in the light verb constructions with suru ‘do’ and the complex adjective constructions formed on the negative adjective nai ‘non-existent.’
1. Japanese as a Nominative-Accusative Language
Natural language makes use of syntactic forms consisting of predicates and arguments to express various semantic relations; even a very simple sentence like John runs comprises a predicate (run) and its argument (John). This coding system of natural language, governed by certain strict grammatical rules, allows the speaker to express any concept or thought in the form of sentences, and at the same time, allows the meanings expressed by the sentences to be understood by the hearer in a consistent and efficient manner.
Nevertheless, the actual coding frames utilized to specify the grammatical relations of arguments vary quite widely, affected by a number of linguistic factors. Arguments are realized in a wide variety of ways even if they bear essentially the same semantic relations to the predicate. Although the grammatical relations of arguments are signaled solely by case marking in Japanese, some intricate behaviors of argument marking are observed, which give clues to uncovering common linguistic features and linguistic universals, and provide data contributing to theories of argument realization (cf. Levin, 1993; Levin & Rappaport Hovav, 2005).
In Japanese, there are at least three types of argument alternation. One type of alternation is trigged purely on the basis of meanings expressed in clauses (e.g., the plural agent and the source agent alternations). A second type of argument alternation is triggered by adding a suffix or an auxiliary—deriving a complex clause from a simple base clause (e.g., causative, indirect passive, and tough-adjective constructions) or by embedding a clause under another predicate (e.g., exceptional case marking constructions). Finally, there is a type of alternation, made available by syntactically motivated operations, which changes the syntactic status of arguments (with no suffixation on the predicate) (e.g., subjectivization and noun incorporation).
This article reports and illustrates a number of representative cases of argument alternations observed in Japanese, which are of relevance to linguistic research falling under the rubric of ‘argument realization’, and it elucidates their unique features, as well as properties that are commonly found cross-linguistically.
1.1. Case Markers in Japanese
Japanese is a nominative-accusative language of the agglutinating type, with the basic word order of SOV (subject–object–verb). Japanese does not implement person/number/gender agreement marking on verbs (except an optional honorific marking on verbs that can be seen as a kind of agreement). In addition, word order is not a good indicator of the grammatical relations of arguments, owing to the fact that the language has a re-ordering operation called ‘scrambling,’ which induces a change in the order of arguments without any change in logical meanings. Thus, in Japanese, postnominal case markers are virtually the sole reliable means to detect the grammatical status of arguments, as well as adjuncts. In discussing argument realizations in Japanese, it is essential to consider how arguments are morphologically marked, because postnominal case markers provide a reliable way to assess the grammatical status of arguments.
Despite the fact that postnominal case marking plays a pivotal role in coding the grammatical relations of arguments, Japanese has only a small number of simple postnominal particles (plus some complex postpositions, perhaps, formed as a result of grammaticalization). Postnominal case markers can be divided into two groups, more or less reflecting the argument or adjunct distinction of nominals. One group of postnominal case markers, which includes nominative ga, accusative o, and dative ni, is often labeled as ‘structural case marker’ or ‘syntactic case marker’; the other group, which includes markers kara ‘from,’ made ‘up to,’ and the like, is referred to as ‘postposition’ or ‘semantic case marker’.
While nominals with syntactic case markers (nominative, accusative, dative) are often amenable to syntactic operations affecting their grammatical relations (such as passivization and causativization), nominals with semantic case markers are not. As noted, these properties are closely correlated with the distinction between argument and adjunct. (Nevertheless, in traditional Japanese grammar, no grammatical distinction of structural case versus semantic case has been posited.) There are many different argument realization patterns observed in Japanese, which are regulated by a number of grammatical rules. Some representative cases will be highlighted in the subsequent discussions.
1.2. Case-Marking Patterns in Basic Clauses
The first discussion centers on the basic patterns of argument realizations in Japanese. Since Japanese belongs to the nominative-accusative language type, it is a general rule that the subject of a transitive verb is treated in the same way as the subject of an intransitive verb, and differently from the direct object of a transitive verb in terms of their case marking.
In transitive clauses, the subject is marked with nominative case and the object with accusative case. In intransitive clauses, the sole argument (i.e., the subject) of the verb is marked with nominative case.
Japanese has ditransitive predicates (selecting three arguments), which take a ‘nominative-dative-accusative’ case-marking pattern, as (3) shows.
Ditransitive predicates most typically describe either a change of possession (e.g., ataeru ‘give’) or a change of location (e.g., okuru ‘send’) (Kishimoto, 2001b). Thus, the dative argument of a ditransitive verb, which functions as an indirect object syntactically, is identified as a recipient or goal.
Transitive predicates selecting subject and object take a nominative-accusative case-marking pattern as an unmarked option, as in illustrated in (2b), but this is by no means the only case pattern available for transitive predicates. Transitive stative predicates often allow variable case-marking patterns. For example, derived transitive stative predicates like uta-e-ru ‘can sing’ (suffixed with the potential marker -e ‘can’), as well as a small set of simple stative verbs like wakaru ‘understand,’ allow their subjects to carry either dative or nominative case (Kuno, 1973; Shibatani, 1978, 2001, and others).
When the subject is marked with nominative case, as in (4a), the object can be marked with either accusative or nominative case. When the subject is marked with dative case, as in (4b), the object can only be marked with nominative case. The unavailability of accusative case marking on the object in (4b) is often attributed to the so-called nominative-case constraint—the requirement that a finite clause must have at least one nominative argument (Shibatani, 1978).
In (2) and (4), the arguments appearing in the initial position (in unmarked word order) are construed as subjects. This fact can be confirmed by appeal to reflexivization—a major subject diagnostic available in Japanese. In the examples in (5), the reflexive zibun ‘self’ invariably takes the initial argument as its antecedent, despite a difference in surface case marking, suggesting that this argument serves as a subject.
The same results are obtained by another subject diagnostic test of subject honorfication. In the examples in (6), the initial arguments are targeted for subject honorification, whether they are marked with nominative or dative case (Harada, 1976; Hasegawa, 2006).
The examples in (6) also show that the initial argument assumes the grammatical function of subject syntactically. The data illustrate that the grammatical relation of the initial arguments in (4) is unvarying regardless of whether they are marked with dative or nominative case.
The facts of potential predicates make it clear that the morphological marking of subjects is not limited to nominative case, but can be dative case, which means that Japanese has so-called dative-subject constructions, which are attested cross-linguistically, in particular, in Asian languages (see e.g., Bhaskararao & Subbarao, 2004). One of the typological issues surrounding the dative-subject construction is how dative case marking on subjects is conditioned. It is generally observed that subjects can be assigned dative case marking when they are construed as experiencers or possessors, and this holds in Japanese, too. In addition, it is often claimed that the semantic meaning of ‘stativity’ plays a crucial role in sanctioning variable case-markings in Japanese (e.g., Koizumi, 2008; Kuno, 1973; Tada, 1992).
While, in Japanese, the subjects of transitive stative predicates are often allowed to bear dative case, it is also true that some transitive non-stative predicates allow the non-canonical dative-nominative case-marking pattern (Kishimoto, 2005). For instance, the verb umareru ‘be born’ can be used as a transitive predicate taking a subject.
In (7), the initial dative argument serves as the subject of the clause. This fact can be confirmed by looking at the examples in (8), which involve reflexivization and subject honorification.
The examples show that the initial dative argument counts as the target of reflexivization and subject honorification. Umareru is a non-stative verb. Nevertheless, when this verb is used as a transitive verb, it takes a dative-nominative case-marking pattern, and the initial dative argument serves as a subject.
It is also worth noting that the verb umareru can have an intransitive use, in which case the sole nominative argument serves as a subject, as exemplified by the fact that reflexive zibun can take the nominative argument as its antecedent.
By contrast, when used as a transitive verb, umareru does not allow its nominative argument to serve as the antecedent of zibun, as (9b) shows. The nominative object of a transitive stative predicate like sikareru ‘can scold’ displays exactly the same behavior in this respect, as (10) illustrates.
It should be clear then that the nominative argument of the transtive umareru does not behave as a subject. The data illustrate that certain transitive non-stative (or eventive) predicates allow the initial dative argument to serve as a subject, and makes the case-marking pattern of dative-nominative available.
Finally, adjuncts typically bear oblique marking; that is, they occur with semantic case markers. Semantic case markers show the same behavior as syntactic case markers in morphological terms, in that they occur after the host nominals. The following examples illustrate how adjuncts are marked.
Caution needs to be exercised with regard to the distinction of argument versus adjunct, because most, if not all, basic case markers are polysemous or homonymous. Thus, the same morphological marker can sometimes signal different grammatical relations of arguments and adjuncts—ni can be used in a number of different ways, such as dative, locative, directional, and goal markers.
Not surprisingly, in Japanese, just like other nominative-accusative languages, nominative case is identified as the most prominent structural case marking. This characterization of prominent case is largely motivated by the fact that in Japanese, the subject, which is required in virtually any kind of finite clause, is typically marked with nominative case.
1.3. Oblique Subjects
Japanese can implement a number of argument alternations, which are reflected on the case marking of arguments. One interesting case is found in subject-marking alternations, which can be implemented when certain semantic conditions are met. In Japanese, subjects in unmarked cases appear with nominative case, but dative case appears on the subjects of certain transitive predicates, most typically transitive potential predicates. Japanese also allow subjects to bear case marking other than nominative and dative marking.
More specifically, in Japanese, a semantic case marker is allowed to occur on subjects (by replacing nominative case) if certain semantic conditions are met (Inoue, 1998; Kishimoto, 2005, 2010; Nitta, 1997). There are at least two kinds of subject case alternations (referred to here as the source subject alternation and the plural agent alternation).
In the source agent alternation, subjects may be assigned ablative kara ‘from,’ in lieu of nominative ga when they are thematically conceived of as representing an agentive source, a starting point (either physically or metaphorically), as in (12a). In the plural agent alternation, the instrumental (or possibly, the locative) de may be substituted for nominative ga on agentive subjects when they refer to a group of people, as in (12b) (Kishimoto, 2005; Takubo, 2010). The oblique kara and de are both semantic case markers used for designating certain specific semantic relations. Importantly, both plural agent and source agent alternations are genuinely motivated on semantic grounds, and do not trigger any morphological change on the predicates.
One important feature of the case-marking alternations at issue is that the initial oblique arguments count as subjects syntactically regardless of whether they are marked with kara or de, as seen from the fact that subject-oriented reflexive zibun ‘self’ can take them as its antecedents.
In Japanese, finite clauses are, more often than not, required to include a nominative argument in them, but the oblique-subject constructions need not comprise any nominative argument, suggesting that they are exempt from the nominative-case constraint. Remarkably, Japanese allows structural case markers on subjects to be replaced by semantic case markers rather freely when certain semantic conditions are satisfied, which does not seem to be commonly found in other languages, including English.
It is occasionally reported that the assignment of non-canonical case marking (other than nominative and dative marking) to subjects is possible in other languages (e.g., Verma & Mohanan, 1990). The important fact regarding the Japanese cases is that, when subjects bear semantic case markers, there is a direct relationship between their marking and their thematic interpretations. This suggests that these alternations are motivated solely on semantic grounds, and thus provide data that allow us to assess how argument realizations are regulated in the mapping from lexical-semantic representations to syntactic structures (Levin & Rappaport Hovav, 2005).
1.4. ECM and Small-Clause Constructions
In Japanese, some case markings that are not available in basic clauses are found in certain embedded contexts. In the exceptional case marking (ECM) construction in (14a) and the small-clause construction in (14b), the embedded subject receives accusative case marking.
The type of case marking found in (14) is not possible with subjects that appear in simple clauses (or in matrix clauses). Notably, the Japanese ECM construction allows the complementizer to ‘that’ to appear in the subordinate clause, although it behaves as a non-finite clause syntactically. (The fact that the ECM construction comprises the complementizer to is peculiar, because this kind of complementizer is normally not found in ECM constructions in other languages.) In the small-clause construction, the embedded predicate appears in the adverbial form (without tense). Verbs that can take ECM complements include omou ‘think,’ sinziru ‘believe,’ and katei-suru ‘assume,’ and verbs that can select small-clause complements include omou ‘think’ and suru ‘make.’
In both ECM and small-clause constructions, the matrix predicate can be assumed to license the accusative case marking on the embedded subjects, given that the accusative-marked arguments in these constructions can be promoted to subjects when direct passivization applies to the upper matrix predicate, as shown in (15).
The accusative case marking on the embedded subjects is naturally expected, however, because the verb omou can take an accusative object.
This fact suggests that in both ECM and small-clause constructions, the predicate in the matrix clause serves as the licenser of accusative case marking. Then, the data in (15) and (16) point to the conclusion that when a stative predicate like kawaii ‘cute’ is embedded under a predicate like omou ‘think,’ the upper predicate assigns accusative case marking to the subject of the embedded predicate.
1.5. Locative Alternation
Just like many other languages in the world, Japanese can instantiate locative alternation—one type of cross-linguistically observed case marking alternation applying to basic clauses. In Japanese, verbs that induce locative alternation express the meanings of ‘filling’ or ‘spreading,’ as in mitasu ‘fill,’ nuru ‘paint,’ tirakasu ‘scatter,’ and kazaru ‘decorate.’ The examples in (17) illustrate the two case frames available for the verb nuru ‘painting.’
In the material-object variant in (17a), the theme receives accusative marking while the location is marked with locative ni. In the location-object variant in (17b), the location is marked with accusative case, and the theme with instrumental de. Verbs with the meaning of ‘removal’ display a different case-marking pattern.
In the material-object in (18a), the theme is marked with accusative case and the source with ablative kara ‘from.’ In the location-object variant in (18b), the source is marked with accusative case, but the theme must be left unexpressed in the absence of an oblique marker that can be used for marking this argument (Kageyama, 1981).
Importantly, the material-object variant in (17a) takes the case frame found with verbs of motion, and the location-object variant in (17b) has the case frame taken typically by verbs of change of state. This fact suggests that locative alternation verbs display two distinct argument realization patterns owing to the distinct meanings they carry (Kishimoto, 2001a). The adequacy of this view is confirmed by the examples in (19), where the locative alternation verb nuru ‘paint’ is compounded with another verb.
In the first place, the second verb tukeru ‘attach,’ which signifies a directional movement, eliminates from the first verb nuru ‘paint’ the component of lexical meaning indicating a change of state. Consequently, only the locative-object variant is allowed for the compound verb nuri-tukeru ‘paint-attach.’ In the second place, if ageru ‘up, finish,’ which denotes a completion of action, is compounded with nuru ‘paint,’ it serves to eliminate the motion meaning from the first verb nuru. Due to the elimination of the motion meaning, only the material-object variant is available for the compound verb nuri-ageru ‘paint-up.’
The data in (19) show that compounding of a locative alternation verb with another verb results in impeding the locative alternation. Since the possibility of locative alternation changes according to what meaning the predicate carries, locative alternation must be conditioned on semantic grounds (cf. Okutsu, 1981). Locative alternation provides a testing ground for the long-standing issue of how arguments are morphologically coded according to the meaning expressed by the verb—that is, the issue of whether there are semantic factors that affect the case marking of arguments (e.g., Jackendoff, 1990). In particular, the facts of Japanese compound verbs that can invoke locative alternation make it clear that the ambiguity of verbs is an important factor governing the possibility of this type of argument alternation.
2. Complex Predicates
Japanese has agglutinative morphology. Owing to the morphological characteristics of the language, many predicative elements occurring after main predicates count as bound morphemes that need to be merged into their host morphologically. By virtue of the fact that multiple predicative heads are coalesced into one morphological entity, very large multiply-headed complex predicates, which are morphologically single units, are easily formed in Japanese.
The complex predicate in (20) is comprised of five predicative elements followed by the tense marker -ta. The entire complex predicate forms a morphologically tight unit, despite the fact that it comprises a number of embedded clauses syntactically.
Complex predicates constitute a unitary complex morphologically, and often display a discrepancy between the morphological and syntactic structures. In point of fact, it is often pointed out (e.g., Matsumoto, 1996) that complex predicate constructions display mono-clausal properties in some respects, but have bi-clausal properties in other respects. Japanese also has periphrastic verb constructions where verbs do not form morphologically tight units. In these constructions, some argument realization patterns, different from those found in the base clauses, are derived by virtue of the addition of auxiliary elements. Some representative cases of complex predicate formation, including causative and indirect passive formation, tough-constructions, and periphrastic resultative constructions, will be discussed.
Complex causative verbs are formed by agglutinating the causative -(s)ase to verbal predicates. Causativization brings about the effect that a causer argument is added to the base clause. The causative constructions display mono-clausal case-marking patterns, despite the fact that they have bi-clausal structures syntactically. Semantically, causative sentences can indicate directive causation and manipulative causation (Shibatani, 1976). This semantic distinction is manifested in the case marking of the causee argument when the main verb is intransitive.
In the directive causative (21a), the causee is marked with dative ni, and in the manipulative causative (21b), it is marked with accusative o. When the main verb is transitive, the semantic distinction of directive versus manipulative causation is not reflected in the surface case marking, since the causee is invariably marked with dative case regardless of whether it expresses manipulative or directive causation.
In (22), the expected accusative marking does not appear on the cause, even when the sentence expresses manipulative causation. The exclusion of the accusative case on the causee with the causativized transitive verb is attributed to the so-called ‘double-o constraint’, requiring that a single clause cannot have two or more o-marked arguments (see e.g., Harada, 1973; Inoue, 1976; Shibatani, 1978).
The causative constructions have complex syntactic structures, which can be easily seen from the fact that the dative as well as the nominative argument can serve as the antecedent of subject-oriented reflexive zibun ‘self,’ as in (23).
Since the target of reflexive zibun is limited to subjects, the fact indicates that the dative-marked causee does not have the status of adjunct, but serves as an argument (i.e., it retains the status as the subject of the embedded predicate). This fact suggests that a causative sentence should have a syntactic structure where the causative verb takes a complement clause, despite the fact that it has a mono-clausal case array.
Passive constructions are formed by adding the passive morpheme -(r)are to the main verb. Japanese has two distinct types of passives: direct passive and indirect passive. In direct passive clauses, the erstwhile subject is demoted to an adjunct, while its original object being promoted to subject, as illustrated in (24).
In direct passive sentences, the agent argument is optionally realized, because it serves as an adjunct syntactically. When realized, it is marked with the dative ni, the ablative kara ‘from,’ or the postposition niyotte ‘by’ (see e.g., Kuroda, 1979; Masuoka, 1987).
By contrast, when an indirect passive clause is formed, a new subject that signifies an affected experiencer is added, with a concomitant change of the nominative marking on the original subject to the dative ni, as shown in (25).
The indirect passive is sometimes referred to as adversative passive as well, owing to an adversative meaning associated with the clause (see Hoshi, 1999).
The two types of passive clauses show some behavioral differences (Howard & Niyekawa-Howard, 1976; Kuno, 1973; Kuroda, 1965). While only the nominative argument can be the antecedent of subject-oriented reflexive zibun ‘self’ in the direct passive clause, the dative as well as the nominative argument can serve as the antecedent of zibun in the indirect passive clause.
The indirect passive, as opposed to the direct passive, allows two distinct arguments to serve as the antecedent of reflexive zibun. This shows that the thematic subject of the main verb is not demoted to an adjunct by indirect passivization. Given that reflexivization has subject orientation, the data suggest that the indirect passive clauses should have bi-clausal structures despite their mono-clausal case morphology.
Tough-constructions, which are formed by embedding a verb under a so-called tough-adjective (e.g., -zurai ‘hard,’ -nikui ‘difficult,’ -yasui ‘easy’), provide another case illustrating an apparent discrepancy between morphology and syntax. In Japanese, tough-predicates, just like passive and causative predicates, form complex predicates that display mono-clausal frames in case-marking terms.
In the tough-clauses comprising a transitive verb, the subject may be marked with nominative ga, dative ni, or nitotte. (When the subject is not marked with nominative case, the object must appear in the nominative case, because of the nominative-case constraint.) Syntactically, the initial experiencer argument counts as the subject, regardless of its case marking, as the subject-oriented zibun can take this argument as its antecedent, as (28) shows.
The two types of tough-sentences convey different meanings. When the tough-clause takes a non-stative case-marking pattern, as in (27a), a tendency or a dispositional meaning is expressed. When the clause takes a stative case-marking pattern, as in (27b), it expresses the sense of the subject’s control (Inoue, 1978).
2.4. Resultative Constructions
Japanese has periphrastic verb constructions, where the verbs do not form morphologically tight units. Periphrastic resultative constructions are formed by making use of the auxiliary verb aru ‘be,’ in combination with the main verb taking the gerundive (-te) form. The resultative constructions have bi-clausal structures syntactically, but display mono-clausal behavior in case-marking terms.
Japanese has two types of resultative constructions—one is a transitivity-maintaining type, and the other, an intransitivizing type (Masuoka, 1987). In the transitivity-maintaining type, a clause is simply embedded under the verb aru, as in (29a), and in the intransitivity type, the case-marking pattern changes, as in (29b).
In the intransitivizing type of the resultative construction in (29b), the main predicate is transitive, but the construction assumes a passive-like function, promoting a theme (i.e., an object) to the subject while defocusing an agent. In this resultative construction, the existence of an agent is clearly implied, because it signifies that an action has been taken for some specific purpose (Matsumoto, 1990; Teramura, 1984). Nonetheless, the resultative construction in (29b), unlike the direct passive construction, never allows the overt realization of an agent syntactically.
Another difference that distinguishes the resultative constructions from direct passive clauses lies in the fact that the subject of the periphrastic resultative construction need not be an argument of the verb embedded directly under aru.
In (30a), the theme argument marked in the nominative case counts as the object of the verb yomu ‘read,’ which appears in the complement clause selected by yuu ‘say.’ (30a) looks like a direct passive clause that promotes an object to the subject of the clause. Nevertheless, it is not possible to construe (30a) as invoking passivization, because direct passivization applying to the verb yuu ‘say’ cannot promote the object of yomu to the subject, as seen in (30b).
3. Grammatically Derived Argument Alternations
Japanese has a number of grammatical processes that change the transitivity of clauses without effecting any morphological changes on the predicates. These are the syntactic operations that invoke changes the number of arguments available in the clause, as well as their syntactic status. Since the meanings of the constructions do not change regardless of whether the operations apply, they are not semantically motivated. In Japanese, there are both grammatical operations increasing the number of arguments and grammatical operations reducing the number of arguments.
One argument-increasing operation is called subjectivization in the literature on Japanese (Kuno, 1973). Many instances of subjectivization can be regarded as involving argument ascension (or possessor raising), which have been observed in many typological studies and analyzed in various theoretical frameworks (including generative grammar). Japanese implements another type of grammatical operation, which is often referred to as noun incorporation. This grammatical process reduces the number of arguments available in the clause, and is also widely attested across languages. In Japanese, noun incorporation takes place in the light verb constructions with verbal nouns combined with suru ‘do’ and complex negative predicates formed on nai ‘non-existent,’ suggesting that noun incorporation is most typically instantiated when the predicate is semantically light.
3.1. Subjectivization and Possessor Raising
While predicates in most cases determine the number of arguments available for the clauses, there are cases where the clauses comprise more arguments than are required by the predicates. In Japanese, non-thematic arguments can be added to the clause by way of so-called subjectivization (Kuno, 1973). The added arguments receive structural case marking, most typically, nominative case, but count as non-thematic, since they are not selected by the predicates.
Specifically, in Japanese, a multiple nominative construction can be created by subjectivization, where two or more nominative-marked arguments occur in a single clause. In the two sentences in (31), non-thematic arguments appear in the initial position, marked in the nominative case.
The type of construction found in (31) is often referred to as the ‘major subject’ construction, since the initial nominative phrase is not a thematic argument (i.e., the major subject), which is generally assumed to be licensed a part-whole relation (or an “aboutness” relation) to the argument on its right.
The two sentences in (31) are often claimed to derive from (32a) and (32b), via the rule of subjectivization, which changes a genitive or a locative phrase into a nominative phrase.
When subjectivization applies to the genitive argument in (32a), the multiple nominative-subject sentence in (31a) is derived. If this rule applies to the initial locative argument in (32b), (31b) is derived. In (31a), the possessor argument is extracted from the thematic subject on its right and is rendered into a verbal argument. Thus, the grammatical process of subjectivization deriving (31a) from (32a) may be construed as an instance of ‘possessor raising’, which is commonly found across languages.
Notably, subjectivization can be applied iteratively in a single clause, as long as the added nominative argument is licensed with an “aboutness” relation to the argument on its right, as seen in (33a).
Subjectivization has the effect of creating an extra (major) subject argument, because the thematic subject retains its subject status even when an extra possessor or locative argument is marked with nominative case by this rule (Kuno, 1973, 1976). Arguably, the iteration of the operation is possible here, because the added argument serves as an adjunct in syntactic terms.
Two remarks are in order. First, non-thematic arguments added by means of subjectiziation are often marked with nominative case, the original case marker being removed, as seen in (31). Nevertheless, there are cases where the original semantic case marker on nominals is retained even after subjectivization.
In (34b), derived from (34a), the extra argument carries the ablative kara alongside the nominative ga, illustrating one peculiar property of the multiple nominative construction.
Second, the grammatical operation of rendering a genitive argument into a clausal argument is not limited to the multiple nominative construction. This grammatical operation can sometimes create non-subject arguments, marked in the dative case, as seen in (35).
In (35b), the argument corresponding to the genitive argument in (35a) is marked with dative case. Since the cognitive meaning expressed by the two variants is the same, and since the dative argument is not selected by the predicate tame ni naru ‘be beneficial,’ it must be the case that the dative argument is created by argument ascension (or possessor raising). Although this fact has not been much discussed in the literature, it is in fact quite easy to find many more cases of argument ascension like (35b), suggesting that Japanese should be a language that makes fairly extensive use of argument-ascension operations (Kishimoto, 2013, 2014).
3.2. The Light Verb Construction
The light verb construction is formed by combining the light verb suru ‘do’ with a noun with an argument structure—the so-called verbal noun (VN) (e.g., benkyoo ‘study,’ hyooka ‘evaluation,’ zyogen ‘advice’). The Japanese light verb construction can have two variants; one variant in which the verb suru takes a VN as its complement marked with accusative case, and the other where the VN is directly combined with suru, as illustrated in (36).
In (36), the main predicative content of the construction comes from the VN that carries its own argument structure, and the verb suru ‘do’ appears to have little semantic content. Thus, the construction in (36a), comprised of a verbal noun and suru, is often considered as the Japanese counterpart of a light verb construction (like John took a look at the book).
In the variant of the light verb construction in (36b), the VN-suru form is identified as forming a compound word, while the VN o suru form in (36a) is a periphrastic phrase comprising two separate elements. This fact suggests that the compound form of the light verb construction be derived as a consequence of noun incorporation. As often discussed in the typological literature (e.g., Baker, 1988), in constructions where incorporation is applicable, two variants are usually found—one that represents a pre-incorporation structure, and the other that is derived by applying incorporation. This generalization holds true of Japanese, since two variants are available in the light verb construction, as in (36a) and (36b).
Note further that the two types of light verb forms give rise to a difference in acceptability when a VN expands to a phrase.
As shown by the ungrammaticality of (37b), the noun phrase suugaku no benkyoo [math GEN study] ‘a study of mathematics’ cannot be directly combined with the light verb suru. But no problem arises when the same phrase appears in the accusative case, as in (37a). While this fact suggests the word status of a compound VN-suru form, the fact of kata-nominalization in (38) points to the opposite conclusion (Kageyama, 1993; Kishimoto, 2006).
When the VN-o suru form is nominalized with the suffixation of the -kaka affix, the accusative marker is replaced by the genitive marker, as in (38a). On the other hand, the nominalized VN-suru form is not acceptable, as shown in (38b). The unavailability of the nominalized compound V-suru form in (38b) suggests that the VN and suru be first generated as syntactically independent elements. If this is the case, it follows that the VN-suru form in (36b) is created syntactically by incorporating the VN to suru.
In the light verb construction with a verbal noun of the transitive type, it is often the case that either the verbal noun or its argument is marked with accusative, as (39a) and (39b) exemplify.
The light verb suru ‘do’ allows either ensoo ‘play’ or beetooben ‘the Beethoven’ to be marked with accusative case, as in (39a) and (39b), but not both, as in (39c). The variant in (39c) is excluded in violation of the ‘double-o constraint,’ requiring that a single clause cannot have more than one o-marked argument.
3.3. The Complex Adjective Construction
The complex adjective construction with the negative adjective nai ‘non-existent, null’ also invokes noun incorporation that reduces the number of arguments selected by the predicate. The complex negative adjective construction is derived by combining the base adjective nai with a certain class of nouns (e.g., abunage ‘danger,’ nukari ‘fault,’ atogusare ‘later concern’) in two different ways, as seen by the fact that the noun part of the complex adjective can appear with or without nominative case marking.
The adjectival base nai ‘non-existent, null’ has the same morphological form as the negative marker nai ‘not.’ The syntactic status of these two expressions differs, however, in that no positive counterpart can be derived by replacing the adjective nai with aru, unlike the negative marker nai. Accordingly, the complex negative predicate yurugi (ga) nai [shaking (NOM) null] ‘unshakable’ does not have an affirmative form, as shown by the unavailability of the expression *yurugi (ga) aru [shaking (NOM) have] ‘shaky (?).’
The two variants of the complex adjective have the same core meaning, but differ as to whether the noun part can expand to a phrase. When the noun is marked with nominative case, it is possible to accommodate a modifying expression, but when the noun appears without case marking, noun modification is not allowed.
The data show that when the component noun is case-marked, the complex adjective has a structure where the noun stands as an independent phrase located outside the adjective nai. On the other hand, when the noun appears without nominative case marking, the entire expression behaves as a compound whose internal structure is invisible to the syntax. This fact suggests that the variant without case marking on a noun is derived via incorporating the noun into the adjective base nai, and thus, the complex adjective formation can be construed as involving noun incorporation.
Note further that, when used in isolation, the adjective nai behaves as a predicate that takes either a dative-nominative or a nominative-nominative case-marking pattern. Thus, the subjects of complex negative adjectives may bear either dative or nominative case marking, when the component noun of the nai-adjective bears nominative case marking, as (42a) illustrates.
Nevertheless, nominative case marking is the only possibility of subject marking when the adjective base nai is combined with the component noun without nominative case marking, as shown in (42b).
When the component noun is not case-marked, as in (42b), the dative marking on the subject is excluded. This follows from the nominative-case constraint requiring that there be at least one nominative argument in a finite clause. If the noun is incorporated to nai, it loses the status of an independent argument. In this case, the complex adjective counts as intransitive, so that the subject can only be marked with nominative case. (Note that the grammatical process of noun incorporation induces a transitivity-reducing effect and differs from the phenomenon referred to as ‘case marker drop’ observed in colloquial speech, where no such effect is observed.)
4. Motivations for Argument Alternations
In Japanese, transitive non-stative predicates have the case-marking frame of nominative-accusative, but transitive stative predicates often have other case alignments, such as nominative-nominative and dative-nominative. Besides, Japanese have other patterns of argument realizations, which, in many cases, involve argument alternations. In Japanese, there are at least three types of argument alternation. One type of alternation is trigged purely on the basis of meanings expressed in clauses. The second type of alternation is triggered by adding suffixes and auxiliary elements, deriving complex clauses from simple ones. The third type of alternation only induces changes on the syntactic status of arguments (without changing the clause type).
In the type of argument realization triggered purely on semantic grounds, it is often the case that arguments receive either a syntactic or semantic case marker. This type of argument alternation can take place with no essential semantic difference in meaning (as in the plural agent alternation and the source-agent alternations). There are also cases where case-marking alternations are conditioned by the meanings of their predicates, although the predicate forms are not altered (as in locative alternation).
When alternations are invoked with the addition of verbal suffixes, complex predicates (counting as single morphological units) are derived (e.g., passive, causative, and tough-constructions). In complex predicate constructions, argument marking follows the mono-clausal case array, even when they have multi-clausal clause syntactic structures. Their case-marking patterns are thus often different from those regulated by the base predicates. There are cases where argument realization patterns are altered by adding an auxiliary verb to derive a complex clause (e.g., peripheral resultative constructions), as well as cases where arguments receive different case marking by virtue of embedding a clause under another predicate (e.g., ECM and small-clause constructions).
Variable case marking options are also made available by purely syntactically grammatical operations. Japanese has the grammatical process of subjectivization (many instances of which can be conceived of as involving possessor raising), creating extra (non-thematic) major subjects, as well as the grammatical operation of noun incorporation that reduces the number of arguments, as found in the light verb construction with suru ‘do’ and the complex adjective construction with the negative adjective nai ‘non-existent.’ Subjectivization and noun incorporation alter the argument realization patterns by increasing or decreasing the number of arguments available in the clauses with no suffixation (and also without changing the meanings of the predicates).
5. Critical Analysis of Scholarship
Japanese is one of the most extensively studied languages, and a very large body of work on Japanese grammar is available, many of which fall within the rubric of argument realization research. In Japanese linguistics, there is a long-standing domestic tradition called Kokugogaku (literally, national language study). In the 1970s, a different trend of grammatical research on Japanese grammar emerged, influenced by Western linguistic tradition—the mainstream research conducted along this line is couched by the framework of generative grammar and hence is often referred to as “Japanese generative grammar.”
In both traditional Kokugogaku and Japanese generative grammar, there have been controversies over the status of subjects. In the domestic tradition of Kokugogaku, some studies focus on the grammatical relations of arguments with nominative case marking, and their analyses are largely based on morphological criteria. Since nominative case is a typical marker of subjects, grammatical studies in Kokugogaku tradition often take arguments to be realized subjects when marked with nominative case, explicitly or implicitly (e.g., Hashimoto, 1969; Suzuki, 1975, 1992; Onoe, 1997–1998; Niwa, 2004). There are also grammatical studies that do not assign special status to subjects (e.g., Watanabe, 1971) or deny the relevance of the notion of subject to Japanese grammar (e.g., Mikami, 1953; Kitahara, 1981). With the advent of Japanese generative grammar from the late 1960s to the 1970s, distributional criteria are often used to identify the grammatical relations of arguments, leading to the discovery that objects, as well as subjects, can be marked with nominative case (e.g., Kuno, 1973; Kuroda, 1978; Takezawa, 1987).
The treatment of case marking has undergone changes along the development of modern linguistic theories. This trend has been reflected in studies on argument realizations in Japanese, which have been conducted on diverse linguistic phenomena. From the 1970s to the 1980s, the passive and causative constructions, as well as the tough-constructions, whose predicates form morphologically tight units, were much debated within the framework of Japanese generative grammar (e.g., Kuno, 1973; Inoue, 1976; Kuroda, 1987; Shibatani, 1976, 1978; Sugioka, 1986; Terada, 1990; Kitagawa, 1994). In addition, there are topics, such as passivization (e.g., Washio, 1989–1990; Hoshi, 1994; Ishizuka, 2012), the ECM constructions (e.g., Kuno, 1976; Tanaka, 2002), the potential constructions (Takano, 2003; Bobaljik & Wurmbrand, 2007) and subjectivization (or possessor raising) (e.g., Kuno, 1973; Kishimoto, 2013), that have inspired Japanese linguists’ continuing interests for decades. Furthermore, in recent years, the topics of argument realizations are expanded over a quite wide range of constructions, including the small-clause constructions (e.g., Kikuchi & Takahashi, 1991; Takezawa, 1998), compound verbs (e.g., Kageyama, 1993; Yumoto, 2005), the light verb constructions (e.g., Grimshaw & Mester, 1988; Saito & Hoshi, 2000), and adjectival predicates instantiating noun incorporation (e.g., Kageyama, 1993; Kishimoto & Booij, 2014). Needless to say, how case marking is morphologically realized on arguments is one of the major concerns in all of the grammatical studies on argument realizations in Japanese.
Hinds, J., & Howard, I. (Eds.). (1978). Problems in Japanese syntax and semantics. Tokyo: Kaitakusha.Find this resource:
Kishimoto, H. (2016). Valency and case alternations in Japanese. In T. Kageyama & W. Jacobsen (Eds.), Transitivity and valency alternation: Studies on Japanese and beyond, 125–154. Berlin: De Gruyter Mouton.Find this resource:
Kishimoto, H., Kageyama, T., & Sasaki, K. (2015). Valency classes in Japanese. In B. Comrie & A. Malchukov (Eds.), Valency classes: A comparative handbook, Vol. 1: Introducing the framework, and case studies from Africa and Eurasia (pp. 765–805). Berlin: De Gruyter Mouton.Find this resource:
Kuno, S. (1973). The structure of the Japanese language. Cambridge, MA: MIT Press.Find this resource:
Martin, S. E. (1975). A reference grammar of Japanese. New Haven, CT: Yale University Press.Find this resource:
Masukoka, T. (2000). Nihongo bunpō no shosō [Aspects of Japanese grammar]. Tokyo: Kurosio Publishers.Find this resource:
Miyagawa, S. (1989). Syntax and semantics 22: Structure and case marking in Japanese. San Diego, CA: Academic Press.Find this resource:
Miyagawa, S., & Saito, M. (Eds.). (2008). The Oxford handbook of Japanese linguistics. Oxford: Oxford University Press.Find this resource:
Shibatani, M. (Ed.). (1976). Syntax and semantics 5: Japanese generative grammar. New York: Academic Press.Find this resource:
Shibatani, M. (1990). Languages of Japan. Cambridge, U.K.: Cambridge University Press.Find this resource:
Tsujimura, N. (Ed.). (1999). The handbook of Japanese linguistics. Malden, MA: Blackwell.Find this resource:
Baker, M. (1988). Incorporation: A theory of grammatical function changing. Chicago: University of Chicago Press.Find this resource:
Bhaskararao, P., & Subbarao, K. V. (Eds.). (2004). Non-nominative subjects (2 vols.). Amsterdam: John Benjamins.Find this resource:
Bobaljik, J., & Wurmbrand, S. (2007). Complex predicates, aspect, and anti-reconstruction. Journal of East Asian Linguistics, 16, 27–42.Find this resource:
Grimshaw, J., & Mester, A. (1988). Light verbs and θ-marking. Linguistic Inquiry, 19, 205–232.Find this resource:
Hashimoto, S. (1969). Joshi-jodōshi no kenkyū [Studies on particles and auxiliary verbs]. Tokyo: Iwanami Shoten.Find this resource:
Harada, S. I. (1973). Counter-equi NP deletion. Annual Bulletin, Research Institute of Logopedics and Phoniatrics, 7, 113–147. University of Tokyo.Find this resource:
Harada, S. I. (1976). Honorifics. In M. Shibatani (Ed.), Syntax and semantics 5: Japanese generative grammar (pp. 499–561). New York: Academic Press.Find this resource:
Hasegawa, N. (2006). Honorifics. In M. Everaert & H. van Riemsdijk (Eds.), The Blackwell companion to syntax2 (pp. 493–543). Oxford: Blackwell.Find this resource:
Hoshi, H. (1994). Theta-role assignment, passivization, and excorporation. Journal of East Asian Linguistics, 3, 147–148.Find this resource:
Hoshi, H. (1999). Passives. In N. Tsujimura (Ed.), The handbook of Japanese linguistics (pp. 191–235). Malden, MA: Blackwell.Find this resource:
Howard, I., & Niyekawa-Howard, A. (1976). Passivization. In M. Shibatani (Ed.), Syntax and semantics 5: Japanese generative grammar (pp. 201–237). New York: Academic Press.Find this resource:
Inoue, K. (1976). Henkei bunpō to nihongo [Transformational grammar and Japanese]. Tokyo: Taishūkan.Find this resource:
Inoue, K. (1978). ‘Tough sentences’ in Japanese. In J. Hinds & I. Howard (Eds.), Problems in Japanese syntax and semantics (pp. 122–154). Tokyo: Kaitakusha.Find this resource:
Inoue, K. (1998). Sentences without nominative subjects in Japanese. Center of excellence research report (2A): Researching and verifying an advanced theory of human language, 1–34. Kanda University of International Studies.Find this resource:
Ishizuka, T. (2012). The passive in Japanese: A cartographic minimalist approach. Amsterdam: John Benjamins.Find this resource:
Jackendoff, R. (1990). Semantic structures. Cambridge, MA: MIT Press.Find this resource:
Kageyama, T. (1981). The role of thematic relations in the spray paint hypallage. Papers in Japanese Linguistics, 7, 35–64.Find this resource:
Kageyama, T. (1993). Bunpō to gokeisei [Grammar and word formation]. Tokyo: Hituzi Syobo.Find this resource:
Kikuchi, A., & Takahashi, D. (1991). Agreement and small clauses. In H. Nakajima & S. Tonoike (Eds.), Topics in small clauses (pp. 75–105). Tokyo: Kurosio Publishers.Find this resource:
Kitahara, Y. (1981). Nihongo no bunpō [Japanese grammar]. Tyoko: Chuō Kōron-sha.Find this resource:
Kishimoto, H. (2001a). Locative alternation in Japanese: A case study in the interaction between syntax and lexical semantics. Journal of Japanese Linguistics, 17, 59–81.Find this resource:
Kishimoto, H. (2001b). The role of lexical meanings in argument encoding: Double object verbs in Japanese. Gengo Kenkyu, 120, 35–65.Find this resource:
Kishimoto, H. (2005). Tōgokōzō to bunpōkankei [Syntactic structures and grammatical relations]. Tokyo: Kurosio Publishers.Find this resource:
Kishimoto, H. (2006). Japanese syntactic nominalization and VP-internal syntax. Lingua, 116, 771–810.Find this resource:
Kishimoto, H. (2010). Subjects and constituent structure in Japanese. Linguistics, 48, 629–670.Find this resource:
Kishimoto, H. (2012). Subject honorification and the position of subjects in Japanese. Journal of East Asian Linguistics, 21, 1–41.Find this resource:
Kishimoto, H. (2013). Covert possessor raising in Japanese. Natural Language & Linguistic Theory, 31, 161–205.Find this resource:
Kishimoto, H. (2014). Dative/genitive subjects in Japanese: A comparative perspective. In M. Saito (Ed.), Japanese syntax in comparative perspective (pp. 228–274). New York: Oxford University Press.Find this resource:
Kishimoto, H., & Booij, G. (2014). Complex negative adjectives in Japanese: The relation between syntactic and morphological constructions. Word Structure, 7, 55–87.Find this resource:
Kitagawa, Y. (1994). Subjects in Japanese and English. New York: Garland.Find this resource:
Kitahara, Y. (1981). Nihongo no bunpō [Japanese Grammar]. Tokyo: Chuō Kōron.Find this resource:
Koizumi, M. (2008). Nominative object. In S. Miyagawa & M. Saito (Eds.), The Oxford handbook of Japanese linguistics (pp. 141–164). New York: Oxford University Press.Find this resource:
Kuno, S. (1973). The structure of the Japanese language. Cambridge, MA: MIT Press.Find this resource:
Kuno, S. (1976). Subject raising. In M. Shibatani (Ed.), Syntax and semantics 5: Japanese generative grammar (pp. 17–49). New York: Academic Press.Find this resource:
Kuroda, S.-Y. (1965). Generative grammatical studies in the Japanese language (PhD diss.), Massachusetts Institute of Technology, Cambridge, MA.Find this resource:
Kuroda, S.-Y. (1978). Case-marking, canonical sentence patterns, and counter Equi in Japanese (A preliminary survey). In J. Hinds & I. Howard (Eds.), Problems in Japanese syntax and semantics (pp. 30–51). Tokyo: Kaitakusha.Find this resource:
Kuroda, S.-Y. (1979). On Japanese passives. In G. Bedell, E. Kobayashi, & M. Muraki (Eds.), Explorations in linguistics: Papers in honor of Kazuko Inoue (pp. 305–347). Tokyo: Kenkyusha.Find this resource:
Kuroda, S.-Y. (1987). Movement of noun phrases in Japanese. In T. Imai & M. Saito (Eds.), Issues in Japanese linguistics (pp. 229–271). Dordrecht, Netherlands: Foris.Find this resource:
Levin, B. (1993). English verb classes and alternations: A preliminary investigation. Chicago: University of Chicago Press.Find this resource:
Levin, B., & Rappaport Hovav, M. (2005). Argument realization. Cambridge, U.K.: Cambridge University Press.Find this resource:
Masuoka, T. (1987). Meidai no bunpō: Nihongo bunpō josetsu [The grammar of proposition: Prolegomena to Japanese grammar]. Tokyo: Kurosio Publishers.Find this resource:
Matsumoto, Y. (1990). Constraints on the ‘intransitivizing’ resultative -te aru construction in Japanese. In H. Hoji (Ed.), Japanese and Korean linguistics (pp. 269–283). Stanford, CA: CSLI.Find this resource:
Matsumoto, Y. (1996). Complex predicates in Japanese: A syntactic and semantic study of the notion ‘word’. Stanford, CA: CSLI and Tokyo: Kurosio Publishers.Find this resource:
Mikami, A. (1953). Gendai gohō josetsu [Prolegomena to modern Japanese grammar]. Tokyo: Tōe Shoin.Find this resource:
Nitta, Yoshio. (1997). Nihongo bunpō kenkyū josetsu [Prolegomena to the study of Japanese grammar]. Tokyo: Kurosio Publishers.Find this resource:
Niwa, T. (2004). Shugo to daimokugo [Subject and topic]. In O. Keisuke (Ed.), Asakura nihongo kooza 6: Bunpō II [Asakura Japanese lectures: Grammar II] (pp. 255–278). Tokyo: Asakura Shoten.Find this resource:
Okutsu, K. (1981). Idō-henka-dōshibun [Sentences with motion-change verbs]. Kokugogaku, 127, 21–32.Find this resource:
Onoe, K. (1997–1998). Bunpō o kangaeru: Shugo 1–4 [Thinking about grammar: Subjects 1–4]. Nihongogaku, 16(11), 91–97; 16(12), 88–94; 17(1), 87–94; 17(4), 96–103.Find this resource:
Saito, M., & Hoshi, H. (2000). Japanese light verb constructions and the minimalist program. In R. Martin, D. Michaels, & J. Uriagereka (Eds.), Step by step: Essays on minimalist syntax in honor of Howard Lasnik (pp. 261–295). Cambridge, MA: MIT Press.Find this resource:
Shibatani, M. (1976). The grammar of causative constructions: A conspectus. In M. Shibatani (Ed.), Syntax and Semantics 6: The grammar of causative constructions. New York: Academic Press.Find this resource:
Shibatani, M. (1978). Nihongo no bunseki [An analysis of Japanese]. Tokyo: Taishūkan.Find this resource:
Shibatani, M. (2001). Non-canonical constructions in Japanese. In A. Aikhenvald, R. M. W. Dixon, & M. Onishi (Eds.), Non-canonical marking of subjects and objects (pp. 307–354. Amsterdam: John Benjamins.Find this resource:
Sugioka, Y. (1986). Interaction of derivational morphology and syntax in Japanese and English. New York: Garland.Find this resource:
Suzuki, S. (1975). Shugoron no mondaiten [Problems on subjects]. Gengo, 4(3), 20–27.Find this resource:
Suzuki, S. (1992). Shugoron o megut-te [Issues surrounding subjects]. In Gengogaku Kenkyūkai (Ed.), Kotoba no kagaku5 [Speech sciences 5] (pp. 71–108). Tokyo: Mugi Shobo.Find this resource:
Tada, H. (1992). Nominative objects in Japanese. Journal of Japanese Linguistics 14, 91–108.Find this resource:
Takano, Y. (2003). Nominative objects in Japanese complex predicate constructions: A prolepsis analysis. Natural Language & Linguistic Theory, 21, 779–834.Find this resource:
Takezawa, K. (1987). A configurational approach to case marking in Japanese (PhD diss.), University of Washington, Seattle, WA.Find this resource:
Takezawa, K. (1998). Part I: Kaku no yakuwari to kōzō [The role of case and structure]. In K. Takezawa & J. Whitman, Kaku to gojun to tōgo-kōzō [Case, word order and syntactic structures] (pp. 1–102). Tokyo: Kenkyūsha.Find this resource:
Takubo, Y. (2010). Nihongo no kōzō [The structure of Japanese]. Tokyo: Kurosio Publishers.Find this resource:
Tanaka, H. (2002). Raising to object out of CP. Linguistic Inquiry, 33, 637–652.Find this resource:
Terada, M. (1990). Incorporation and argument structure in Japanese (PhD diss.), University of Massachusetts, Amherst, MA.Find this resource:
Teramura, H. (1984). Nihongo no shintakusu to imi II [Syntax and meaning of Japanese II]. Tokyo: Kurosio Publishers.Find this resource:
Washio, R. (1989–1990). The Japanese passive. The Linguistic Review, 6, 227–263.Find this resource:
Verma, M. K., & Mohanan, K. P. (Eds.). (1990). Experiencer subjects in South Asian languages. Stanford, CA: CSLI.Find this resource:
Watanabe, M. (1971). Kokugo kōbun ron. Tokyo: Haniwa Shobō [On Japanese grammatical constructions].Find this resource:
Yumoto, Y. (2005). Fukugō-dōshi hasei-dōshi no imi to tōgo [The meaning and the syntax of compound and derivational verbs]. Tokyo: Hituzi Syobo.Find this resource: