Case, Argument Structure, and Word Order

Shigeru Miyagawa

• Publisher: Taylor & Francis
• Year: 2012
• Print ISBN: 9780415878593
• eBook ISBN: 9781136458729
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

Over the years, a major strand of Miyagawa's research has been to study how syntax, case marking, and argument structure interact. In particular, Miyagawa's work addresses the nature of the relationship between syntax and argument structure, and how case marking and other phenomena help to elucidate this relationship. In this collection of new and revised pieces, Miyagawa expands and develops new analyses for numeral quantifier stranding, ditransitive constructions, nominative/genitive alternation, "syntactic" analysis of lexical and syntactic causatives, and historical change in the accusative case marking from Old Japanese to Modern Japanese. All of these analyses demonstrate an intimate relation among case marking, argument structure, and word order.

1 Numeral quantifiers and thematic relations¹

1 INTRODUCTION

In Japanese, the act of counting people, animals, or things invariably invokes the use of a numeral quantifier (NQ). A NQ consists of a numeral and a classifier that agrees with the type of entity being counted. For example, to count people, one would use the classifier -nin, as shown in (1). To count bound volumes such as books and magazines, the classifier -satu is used, as shown in (2).²

(1) Sensei-ga san-nin kita.
teacher-NOM three-CL came
‘Three teachers came.’

(2) Hanako-ga hon-o ni-satu katta.
Hanako-NOM book-ACC two-CL bought
‘Hanako bought two books.’

The number of these classifiers is quite large. In fact, there are over 150 classifiers attested, though the inventory of those most commonly used is considerably smaller, possibly less than 30 (Downing 1984, 12–15). When fifteen informants were asked by Downing to choose the classifiers that they use from a list of 154, all chose the 27 listed in (3). They were also asked to list the entities that the classifier can count. Understandably, not all speakers agreed on exactly what entities can be counted with these classifiers. Those entities listed in parentheses are entities that not everyone listed. (Some explanatory notes have also been added in square brackets.)

(3)	1.	dai	furniture, machines, land and air vehicles
	2.	hiki	animals [excluding birds, for some speakers]
	3.	hon	long, slender objects, such as pencils, trees, threads, roads, and lines; (items that follow a trajectory, such as TV programs, letters, telephone calls, and baseball hits)
	4.	kabu	rooted plants, roots and bulbs, shares of stock
	5.	ken	buildings or parts of buildings that act in some functional capacity, such as a home or shop
	6.	ken	incidents, occurrences, such as robberies, fires, and accidents
	7.	ki	airplanes, (other air vehicles, such as helicopters and rockets)
	8.	ko	small objects of roughly equivalent extension in all three dimensions, such as fruits, candies, and stones; also coins [general inanimate classifier, for some speakers]
	9.	ku	haiku [seventeen-syllable poems], (other short poems)
	10.	kyoku	pieces of music
	11.	mai	flat, thin objects such as sheets of paper, pieces of cloth, dishes, items of clothing, phonograph records, leaves, rugs, and coins
	12.	mei	human beings [honorific]
	13.	mon	questions, problems
	14.	mune	buildings
	15.	nin	human beings
	16.	satu	books, magazines, notebooks, books of tickets, pads of paper
	17.	seki	large boats
	18.	soku	pairs of footwear
	19.	soo	small boats
	20.	syoku	meals
	21.	teki	drops of liquid
	22.	ten	points in a score, items in an inventory, works of art
	23.	too	large animals
	24.	toori	methods, opinions
	25.	tu	inanimates, concrete or abstract [general classifier]
	26.	tubu	small, grainlike objects such as grains of rice, grapes, gems, pills, and drops of liquid
	27.	tuu	letters and postcards, documents, (telephone calls)

The fact that NQs occur in Japanese is consistent with what has been observed cross-linguistically. A number of linguists have noted that a NQ system and obligatory plural marking are in complementary distribution: a language can have one of them, but rarely both (Greenberg 1972; Sanches and Slobin 1973; T’sou 1976). Although Japanese has plural markers (-tati, -ra), they are certainly not obligatory because a noun without such marking can typically denote a singular entity or a collection of entities. The word gakusei, for example, can mean ‘student’ or ‘students’. For this reason, the plural markers rarely surface, except when a specific reference to a plurality of entities is required and contextual information alone does not suffice.

For a NQ to function properly, it must of course be construed with the appropriate noun phrase whose referent is being counted. In most cases this is not a problem because the classifier identifies the type of entity being counted. In (4), the NQ yon-satu ‘four _{BOUND VOLUMES} ’ can only be construed with the object NP hon ‘books’ because the classifier -satu is used exclusively to count bound volumes.

(4) Gakusei-ga hon-o yon-satu katta.
student-NOM book-ACC four-CL bought
‘The students bought four books.’

One would not attempt to relate this NQ to the subject, gakusei ‘students’, because ‘student’ is incompatible with the classifier -satu. But if one wishes to count ‘students’ here, it is not enough to simply change the classifier.

(5) ?*Gakusei-ga hon-o yo-nin katta.
student-NOM book-ACC four-CL bought

The classifier -nin is used to count people, hence it should be appropriate with ‘students’. But as shown, this construal is impossible. To make it possible, the NQ must be positioned next to the subject.

(6) Gakusei-ga yo-nin hon-o katta.
student-NOM four-CL book-ACC bought
‘Four students bought books.’

The grammatical examples up to now, (4) and (6), happen to have the NQ adjacent to its NP. Is it the case that adjacency is the condition required? If so, not much more needs to be said. As shown in the following examples, however, successful NP-NQ construals need not always involve adjacency (see Haig 1980 and Kuroda 1980 on examples such as (8)).

(7) Gakusei-ga kyoo san-nin kita.
student-NOM today three-CL came
‘Three students came today.’

(8) Hon-o Taroo-ga ni-satu katta.
book-ACC Taro-NOM two-CL bought
Lit. ‘Books, Taro bought two.’

In (7), the time adverb kyoo ‘today’ intervenes between the subject NP and its NQ. In (8), the object NP has been scrambled to the head of the sentence, resulting in the subject NP intervening between it and its NQ. Examples such as (7) and (8) demonstrate that the NP and its NQ are syntactically independent phrases. Otherwise, an adverb, as in (7), or a subject NP, as in (8), could not possibly intervene.³

In all of the examples that we have seen thus far, the NP and the NQ are separate phrases. There is an alternative construction in which the two occur in the same phrase. As shown in (9a), in this construction the NQ is in the modifier position of the NP headed by the “counted” noun. The construction with the NQ occurring separately is given in (9b) for com...

Table of contents

• Cover
• Full Title
• Copyright
• Routledge Leading Linguists
• Dedication
• Table of Contents
• Introduction
• 1  Numeral quantifiers and thematic relations
• 2  Telicity, stranded numeral quantifiers, and quantifier scope
• 3  Argument structure and ditransitive verbs in Japanese
• 4  Nominalization and argument structure
• 5  Genitive subjects in Altaic and specification of phases
• 6  The genitive of dependent tense in Japanese and its correlation with the genitive of negation in Slavic
• 7  Blocking and Japanese causatives
• 8  Blocking and causatives revisited
• 9  Historical development of the accusative case marker
• 10  The Old Japanese accusative revisited
• Notes
• References
• Index