Semantic Bleaching and Reanalysis in Grammaticalization of yuk ‘go’
September 24, 2014
Fumihito Arai
Contents
1
Objectives
1
2
Semantics-based Assumption:
2.1 Reanalysis of V-te-yuk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2 Remaining questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
2
3
3
Semantic Bleaching and Reanalysis in Grammaticalization
3.1 Semantic bleaching . . . . . . . . . . . . . . . . . . . . .
3.2 Reanalysis . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 Renewal . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4 Association between semantic bleaching and reanalysis
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4
4
5
5
5
Tests in the Literature
4.1 Tests on Complexity of the V1 -te V2 Predicates
4.1.1 Intervention of Particles . . . . . . . . .
4.1.2 Negation of V1 . . . . . . . . . . . . . .
4.1.3 The o-V-ni nar Honorification . . . . . .
4.2 Tests on Simplicity of the V1 -te V2 Predicates .
4.2.1 NPI-licensing . . . . . . . . . . . . . . .
4.2.2 Crossed Scrambling . . . . . . . . . . .
4.2.3 Adjunct Modification . . . . . . . . . .
4.2.4 Phonological Contraction . . . . . . . .
4.3 Section Summary . . . . . . . . . . . . . . . . .
4
5 Testing the morphological integrity of V-te-yuk
5.1 Aims . . . . . . . . . . . . . . . . . . . . . .
5.2 Observation . . . . . . . . . . . . . . . . . .
5.2.1 Intervention of Particles . . . . . . .
5.2.2 Negation of V1 . . . . . . . . . . . .
5.2.3 The o-V-ni nar Honorification . . . .
5.2.4 NPI-licensing . . . . . . . . . . . . .
5.2.5 Crossed Scrambling . . . . . . . . .
5.2.6 Adjunct Modification of V2 . . . . .
5.2.7 Phonological Contraction . . . . . .
5.3 Results of Observation . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7
7
7
7
7
8
8
9
9
9
10
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
10
10
10
11
11
12
13
13
14
14
6 Control Verb vs. Raising Verb
14
References
15
1
Objectives
• To consider which of the following patterns (Table 2) is the most plausible for explaining the
renewal from the infinitival form, V-yuk ‘V-go’, to the -te conjunctive form, V-te-yuk ‘V-Con-go’,
in the grammaticalization process of Japanese yuk ‘go’;
1
• To examine whether or not the reanalysis of V-te-yuk ‘V-Con-go’, as illustrated in (1), is grounded
from a syntactic viewpoint as well.
Table 1: Possible patterns of semantic bleaching and reanalysis in grammaticalization.
Pattern
Semantic bleaching
Renalysis
A
B
C
D
No
No
Yes
Yes
No
Yes
No
Yes
(1) Reanalysis of V-te-yuk ‘V-Con-go’
Before:
After:
[ V-te ] -yuku
V-[ te-yuku ]
(movement after V / movement while V)
(aspect)
• Examples of V-yuk ‘V-go’ and V-te-yuk ‘V-Con-go’:
2
2.1
(2)
a. {Ken-ga / Ressya-ga} {yamamiti-o
/ tekkyoo-o} hasiri-yuk-u kookei
{Ken-Nom / train-Nom} {mountain.road-Acc / iron.bridge} run-go-Pres scene
‘The scene of {Ken running along a mountain road / a train passing a railroad bridge}.’
b. Hana-ga
kare-yuk-u kookei / Aki-ga
fukamari-yuk-u koro
flower-Nom die-go-Pres scene / autumn-Nom deepen-go-Pres time
‘The scene of a flower dying / The time when autumn grows deeper’
(3)
a. {Ken-ga / Ressya-ga} {yamamiti-o
/ tekkyooo-o} hasit-te-yuk-u kookei
{Ken-Nom / train-Nom} {mountain.road-Acc / iron.bridge} run-Con-go-Pres scene
‘The scene of {Ken running along a mountain road / a train passing a railroad bridge}.’
b. Hana-ga
kare-te-yuk-u kookei / Aki-ga
fukamat-te-yuk-u koro
flower-Nom die-Con-go-Pres scene / autumn-Nom deepen-go-Pres time
‘The scene of a flower dying / The time when autumn grows deeper’
Semantics-based Assumption:
Reanalysis of V-te-yuk
• I propose that the reanalysis of V-te-yuk ‘V-Con-go’ as in (1) prompted the renewal from V-yuk
‘V-go’ to V-te-yuk ‘V-Con-go’; consequently, V-te-yuk came to have similar meanings to V-yuk
and the infinitival form has become less productive.
• See (1). Before the reanalysis, the boundary lies between -te and -yuk under the assumption that
-yuk in this case behaves more like a lexical verb. After the reanalysis, on the other hand, -te
and -yuk form a morphological unit as a result of both of them being semantically bleached and
-yuk becomes an auxiliary and less autonomous.
• Evidence for the semantic bleaching of te: the role of te changes from sequencing the events in
chronological order (e1 < e2 ) to connecting the events (e1 = e2 ).
(4)
te’s role: e1 < e2
oki-te-ikaba imo ba ma-kanasi
put-Con-go.if you Top Pref-sad
2
‘I feel sorry for you, my wife, if I leave you here and go away.’
(Man’yˆoshu.14.3567)
ˆ
(5)
te’s role: e1 = e2
wagipye no kadwo wo sugwi-te-yuk-u ramu
my.house Gen gate Acc pass-Con-go-Pres would
‘My sweetheart will go passing the gate of my house. . .’
(Man’yˆosyu.11.2401)
ˆ
(6)
te’s role: e1 = e2
hi-no
tumat-te-yuk-u
sewasinai aki-ni
daytime-Nom shorten-Con-go-Pres busy
autumn-by
‘[It was when everyone was attracted] by the bustle of autumn, with daytime getting
short.’
(Kokoro)
• As a result of the semantic bleaching of te and auxiliarization of -yuk, the semantics of V-te-yuk,
which bears aspectual meaning, becomes equivalent to that of V-yuk (See (7) and (8)). In short,
the semantic equivalence between the two makes the renewal from V-yuk to V-te-yuk happen.
(7) -yuk ‘-go’
[ (aspect)
]
ARG = P-ARG: p, ARG2: VP
Section
Truth-conditional
s < f ,
FORMAL: DIS(p, Loc(e, s′ )) < DIS(p, Loc(e, f )),
QL =
POV(p)<POINT(e) = Loc′ (e, s′ ), VIEW(y) = ⟨se , fe ⟩>
[ CONST: ϕ
]
Non-truth-conditional Section
TELIC:
BEIdent (x, zstate )
(8) te-yuk ‘Con-go’
(aspect)
[
]
ARG = P-ARG: p, ARG2: VP
Section
Truth-conditional
s
<
f
,
FORMAL: DIS(p, Loc(e, s′ )) < DIS(p, Loc(e, f )),
QL =
POV(p)<POINT(e) = Loc′ (e, s′ ), VIEW(y) = ⟨se , fe ⟩>
[ CONST: ϕ
]
Non-truth-conditional Section
TELIC:
BEIdent (x, zstate )
2.2
Remaining questions
• Assumptions drawn from (7) and (8) are:
1. In aspectual usage, te and -yuk can be recognized as one lexical item, -te-yuk.
2. In aspectual usage, both V-yuk and V-te-yuk has the same syntactic structures as in (9ab),
with two arguments, prop (the point-of-view holder) and VP. 1
1
The syntactic structures proposed here are based on Nishigauchi’s (2009, 2014) proposal of POV (point of view)
projections for analyzing the reflexive zibun in Japanese.
3
(9)
b. Syntax of V-te-yuku (aspect)
DeixP
a. Syntax of V-yuku (aspect)
DeixP
HH
HH
H
H
HHH
H
Deix′
Spec
prop
H
HH
VP
HH
NP
V′
(subject)
V
Deix′
Spec
Deix
-yuku
V1
HH
H
prop
VP
HH
NP
V′
(subject)
V
Deix
-te-yuk-
V1
• Remaining questions:
1. Is there any evidence which shows -te-yuk is one lexical item in the lexicon?
2. Is the reanalysis of V-te-yuk necessary for explaining the renewal from V-yuk to V-te-yuk?
3
3.1
Semantic Bleaching and Reanalysis in Grammaticalization
Semantic bleaching
• It has been acknowledged, since the beginning of research on grammaticalization (e.g. Gabelentz 1891, Meillet 1912), that grammaticalization involves the loss of semantic content; the
weakening of meanings of a lexical item, together with the loss of phonetic substance and
syntactic freedom, suggests that a lexical item is turning to a grammatical item (Heine & Reh
1984).
• It is obvious that meanings become weakened or bleached in the grammaticalization process;
however, meaning change is not a sudden loss but a shift of meaning: “one meaning is demoted,
another promoted” (Hopper & Traugott 2003:94-96).
• Meaning change is considered a shift, because the original lexical meanings, which become
abstract during grammaticalization, are reflected in meanings of a grammatical form and constrain the distribution of the grammatical form (a phenomenon called “persistence” in Hopper
1991).
• The development (or metaphorical extension) of go-verbs:
– Cross-linguistically, the original sense of physical movement of go-verbs becomes abstract
and their grammatical forms express continuative or future meaning (Sweetser 1988, Heine,
Claudi, & Hunnemeyer
1991, Lichtenberk 1991).
¨
– Such a semantic shift (i.e. space > time) is common when spatial terms evolve into temporal
particles and auxiliaries (e.g. Bybee, Perkins, & Pagliuca 1994).
– Point of view of go-verbs: the increase of subjectivity when go-verbs are used as an auxiliary.
As a lexical verb, the directionality of go is anchored in both the subject’s and the speaker’s
point of view, while as an auxiliary it is only anchored in the speaker’s subjective point of
view (Langacker 1990, Hopper & Traugott 2003).
• In contrast, Roberts and Roussou (2003) and Roberts (2010) propose that what is lost in semantic
bleaching is non-logical meaning of an lexical item and logical meaning remains intact in the
process (See also von Fintel 1995).
4
3.2
Reanalysis
• Reanalysis is defined as “change in the structure of an expression or class of expressions that does
not involve any immediate or intrinsic modification of its surface manifestation” (Langacker
1977:58), and thus involves a change in constituency, hierarchical structure, category labels,
grammatical relations, and cohesion (Harris & Campbell 1995:61).
• From the minimalist framework (?, Roberts 2010), reanalysis tends to be upward in the syntactic
structure, and involves:
1. the suppression of a movement relation
2. the formerly-moved element is first-merged in a higher position than previously in the
syntactic structure.
(Roberts 2010:52)
• The upward reanalysis suggests that lexical verbs become functional elements, with the loss of
argument structure, and that its upward path corresponds to Cinque’s (1999, 2004) hierarchy of
functional heads.
3.3
Renewal
• Renewal, a process whereby existing meanings take on new forms (which are often periphrastic),
has been observed in a variety of phenomenon (See Hopper & Traugott 2003 and references
therein).
• It is a historical process which has already identified as early as genesis of the grammaticalization
studies (e.g. Gabelentz 1891, Meillet 1912).
• According to Hopper (1991), the overlap or layering between older and newer expressions at a
given synchronic stage is an indication of the renewal process.
• An example of renewal: the development of Latin and French future tense forms of we will sing
(10)
Pre-Latin
bh umos
*kanta
sing be-2Pl.pres.
Latin
>
French
canta-bimus
sing-2Pl.fut.
cantare habemus
>
sing have-2Pl.pres.
chante-rons
sing-2Pl.fut.
allons chanter
> ?
go-2Pl.pres sing
(Eckardt 2007)
• Such a cyclic historical transition has widely been observed in the emergence of tense or aspect
forms, according to Smith (2006).
3.4
Association between semantic bleaching and reanalysis
• A question that arises: whether or not semantic bleaching and reanalysis are mutually (in)dependent
in the grammaticalization process of Japanese yuk ‘go’; for instance, semantic bleaching always
underlies reanalysis, or the former has nothing to do with the latter, or vice versa.
• I take Pattern D as a working hypothesis.
5
Table 2: Association between semantic bleaching and reanalysis in grammaticalization.
Pattern
Semantic bleaching
Renalysis
A
B
C
D
No
No
Yes
Yes
No
Yes
No
Yes
• Patterns A and B can be ruled out. Grammaticalization cannot take place by these two patterns
because a lexical item does not grammaticalize without semantic bleaching; lexical meaning
remains intact so that a grammatical form never arises. Following Longobardi (2001) and
Roberts and Roussou (2003), reanalysis (i.e. syntactic change) should not arise unless it is
motivated by other types of change in a lexical item (e.g. phonological changes and semantic
changes).
• Patterns C and D will be disputable for the present study. A crucial difference between these
two patterns is whether or not reanalysis is postulated.
– In the literature on grammaticalization, a number of researchers agree with the view that
grammaticalization is manifested in reanalyses (Heine & Reh 1984, Heine et al. 1991, Hopper & Traugott 2003, among others); Hopper and Traugott (2003:69) states that “reanalysis
is the dominant mechanism driving it [grammaticalization].”
– In contrast, Haspelmath (1998) and De Smet (2012) challenge this mainstream view;
Haspelmath (1998) argues that various syntactic changes instantiate grammaticalization
but should not be explained in terms of reanalysis, saying that “Grammaticalization and
reanalysis are disjoint classes of phenomena” (p. 315).
• As for Japanese yuk ‘go’, Miyagawa (1987) proposes restructuring for the verb when it appears
as, what he calls, the Purpose Expression, which is exemplified in (11).
(11) [Taroo-ga [PRO hon-o
kai-ni] it-ta].
Taro-Nom
book-Acc buy-for go-Past
‘Taro went to buy a book.’
(Miyagawa 1987:273)
(12)
a. Taroo-ga Kanda-ni hon-o
zitensya-de kai-ni it-ta.
Taro-Nom Kanda-to book-Acc bicycle-by buy-for go-Past
‘Taro went to Kanda by bicycle to buy a book.’
b. *Taroo-ga hon-o
zitensya-de kai-ni Kanda-ni it-ta.
Taro-Nom book-Acc bicycle-by buy-for Kanda-to go-Past
(Miyagawa 1987:275)
(13) [S NP [VP [S′ [S NP [VP NP kai-ni ] ] ] ik ] ]
→ [S NP [VP NP [V kai-ni ik ] ] ]
• A question remains: whether or not restructuring proposed by Miyagawa (1987) takes place in
V-te-yuk ‘V-Con-go’ as well (the -te conjunctive form is not examined in his paper).
• Although V-te-yuk ‘V-Con-go’ is a different construction from the purpose expression, V-ni ik
‘V-for go’, the -te conjunctive form too may undergoes restructuring because the same verb is
used in these constructions.
6
• Additionally, restructuring is limited to a portion of verbs, as Cinque (2001:47) questions: “[. . .]
why it should exist at all, and why it should exist with those particular verb classes (modal,
aspectual and motion).”
• In light of these arguments, I taking Pattern D as a working hypothesis.
4
Tests in the Literature
4.1
Tests on Complexity of the V1 -te V2 Predicates
(14)
4.1.1
a. Particles can intervene between -te and V2 (Martin 1975:510ff).
b. V1 can only be negated (Nakatani 2013:109).
c. The o-V-ni nar honorification is not allowed (Kageyama 1993:360).
Intervention of Particles
• The fact that particles can intervene between -te and V2 suggests a morphological boundary
between them:
(15) Taroo-wa sono hon-o
gakkoo-ni mot-te-wa-it-ta
ga, . . .
Taro-Top the book-Acc school-Dat have-Ptcle-go-Past but
‘Taro did bring the book to school, but . . .’
(Matsumoto 1996:242)
(16)
a. Kare-wa nanimo kat-te-wa
ko-nakat-ta.
he-Top anything buy-Con-Ptcl come-Neg-Past
‘He did not buy anything (but did something, such as checking the price tags).’
b. Watasi-wa sonna koto omot-te-mo
mi-nakat-ta.
I-Top
such thing think-Con-Ptcl see-Neg-Past
‘I didn’t even think about it.’
(Nakatani 2013:109)
4.1.2
Negation of V1
• That V1 can be negated also indicates the complex aspect of the V1 -te V2 predicates:
(17)
a. Taroo-wa atoato-no tameni
teki-o
tukura-nai-de oi-ta.
Taro-Top future-Gen for.the.sake enemy-Acc create-Neg-Con put-Past
‘For the sake of the future, Taro didn’t make enemies.’
b. Taroo-ga Ziroo-o sikara-nai-de yat-ta.
Taro-Nom Jiro-Acc scold-Neg-Con give-Past
‘Taro didn’t scold Jiro (to Jiro’s benefit).’
(Nakatani 2013:109)
4.1.3
The o-V-ni nar Honorification
• Impossibility of the o-V-ni nar honorification shows that the V1 -te V2 predicates are so-called
‘syntactic compound’, not a lexical compound (Kageyama 1993:360):
(18)
a. Yamada-sensee-ga
sore-o tabe-te mi-ta.
Yamada-teacher-Nom that-Acc eat-Con see-Past
‘Professor Yamada tried eating it.’
7
b. *Yamada-sensee-ga sore-o o-tabe-te mi-ni nat-ta.
(cf. Yamada-sensee-ga sore-o o-tabe hazime-ni nat-ta.)
(Nakatani 2013:110)
4.2
Tests on Simplicity of the V1 -te V2 Predicates
(19)
4.2.1
a. Syntactic evidence:
i. NPI-licensing across the teP boundary is allowed when the te-clause is concatenated
(McCawley & Momoi 1986, Matsumoto 1996, and others).
ii. Internal arguments of an adjunct clause and the matrix clause can be cross-scrambled
in the concatenated cases (McCawley & Momoi 1986, ?).
iii. In concatenated cases, an adjunct modifier cannot selectively modify V2 (McCawley
& Momoi 1986, Matsumoto 1996).
b. Phonological evidence: contraction occurs if V2 begins with a vowel (Nakatani 2013:115).
c. Psycholinguistic evidence (Nakatani 2006, 2013)
NPI-licensing
• In general, a negative polarity item (NPI) must be locally licensed by a Neg head. An adjunct
clause is an island: a NPI in an adjunct clause cannot be licensed by a Neg outside the adjunct
clause.
(20)
a. Taroo-wa [ nanimo hosiku-nakat-ta kara
] sat-ta.
Taro-Top [ anything want-Neg-Past because ] leave-Past
‘Taro left because he didn’t want anything.’
b. *Taroo-wa [ nanimo hosikat-ta kara
] sara-nakat-ta.
Taro-Top [ anything want-Past because ] leave-Neg-Past
(Nakatani 2013:111)
• According to Nakatani (2013), te phrases constitute a barrier for the NPI-licensing in the nonconcatenated cases; thus, (21b) is ungrammatical in contrast to (21a). However, the NPIlicensing across te phrases become grammatical in the concatenated cases like (21c).
(21)
a. Taroo-wa [ dokonimo ika-nai-de ] ringo-o tabe-ta.
Taro-Top [ anywhere go-Neg-Te ] apple-Acc eat-Past
‘Taro didn’t go anywhere and ate an apple.’
b. *Taroo-wa [ dokonimo it-te ] ringo-o tabe-nakat-ta.
Taro-Top [ anywhere go-Te ] apple-Acc eat-Neg-Past
c. Boku-wa nanimo tabe-te ko-nakat-ta.
I-Top
anything eat-Te come-Neg-Past
(Nakatani 2013:111)
• In short, that the NPI-licensing across te phrases is possible means that in such a case the V1 -te
V2 behaves as a single predicate.
8
4.2.2
Crossed Scrambling
• When the internal arguments of V1 and V2 can be cross-scrambled, the V1 -te V2 predicate seems
to behave like a single predicate (Nakatani 2013:14).
• In general, a sentence becomes ungrammatical when the internal arguments of an adjunct
clause and the matrix clause cross-scrambled. In the non-concatenated case like (22a-c), the
internal arguments of the te phrase and the matrix ik, hon-o ‘book-Acc’ and gakkoo-ni ‘schoolDat’ respectively, cannot be cross-scrambled, as in (22c).
(22)
a. Taroo-wa [ hon-o
mot-te ] zitensya-de gakkoo-ni it-ta.
Taro-Top [ book-Acc hold-Te ] bicycle-with school-Dat go-Past
‘Taro went to school by bicycle, holding a book in his hand.’
b. Taroo-wa gakkoo-ni [ hon-o
mot-te ] zitensya-de it-ta.
Taro-Top school-Dat [ book-Acc hold-Te ] bicycle-with go-Past
c. *Taroo-wa [ hon-o
gakkoo-ni mot-te ] zitensya-de it-ta.
Taro-Top [ book-Acc school-Dat hold-Te ] bicycle-with go-Past
(Nakatani 2013:113)
• In contrast, the fact that the crossed-scrambling is possible in (23a-c) shows that mot-te-ik in
these examples behaves like a single predicate (Nakatani 2013:113).
(23)
a. Taroo-wa zitensya-de gakkoo-ni hon-o
mot-te it-ta.
Taro-Top bicycle-with school-Dat book-Acc hold-Te go-Past
‘Taro brought a book to school with a bicycle.’
b. Taroo-wa zitensya-de hon-o
gakkoo-ni mot-te it-ta.
Taro-Top bicycle-with book-Acc school-Dat hold-Te go-Past
c. Taroo-wa hon-o
zitensya-de gakkoo-ni mot-te it-ta.
Taro-Top book-Acc bicycle-with school-Dat hold-Te go-Past
(Nakatani 2013:113)
4.2.3
Adjunct Modification
(24)
Adjunct modification:
a. Tanaka-san-wa syorui-o
mot-te faasuto-kurasu-de ki-ta.
Tanaka-Mr-Top document-Acc hold-Te first-class-with
come-Past
‘Mr. Tanaka held the documents and came using a first-class ticket.’
b. ?Tanaka-san-wa faasuto-kurasu-de syorui-o
mot-te ki-ta.
Tanaka-Mr-Top first-class-with
document-Acc hold-Te come-Past
c. Tanaka-san-wa kuruma-de syorui-o
mot-te ki-ta.
Tanaka-Mr-Top car-with
document-Acc hold-Te come-Past
‘Mr. Tanaka brought the documents by car.’
(Nakatani 2013:114)
4.2.4
Phonological Contraction
(25)
Contraction:
a. Kare-wa nanimo mot-te
ika-nakat-ta.
he-Top anything hold-Con go-Neg-Past
‘He didn’t bring anything.’
b. Kare-wa nanimo mot-te’ka-nakat-ta.
9
4.3
Section Summary
• The result of the tests discussed above is summarized as follows (Tables 3 and 4).
Table 3: Tests of complexity.
Particle intervention
V1 negation
o-V-ni nar honorification
5
Table 4: Tests of simplicity.
Ok
Ok
No
NPT-licensing
Crossed-scrambling
V2 modification
Contraction
Ok
Ok
No
Ok
Testing the morphological integrity of V-te-yuk
5.1
Aims
• I use the tests outlined in Section 4 to show that the reanalyzed -te-yuk is morphologically
unified (i.e. V-[te-yuk]) by comparison with -te-yuk denoting movement, the one before being
reanalyzed (i.e. [V-te]-yuk). I expect to obtain the following results (Tables 5 and 6).
Table 5: The expected result as to the tests on complexity.
Before reanalysis: [V-te]-yuk
After reanalysis: V-[te-yuk]
Particle intervention
V1 negation
o-V-ni nar honorification
Ok
Ok
No
No
No
Ok
Table 6: The expected result as to the tests on simplicity.
Before reanalysis: [V-te]-yuk
After reanalysis: V-[te-yuk]
NPI-licensing
Crossed-scrambling
V2 modification
Contraction
5.2
No
No
Ok
Unfavorable
Ok
Ok
No
Favorable
Observation
5.2.1
Intervention of Particles
• A variety of particles which can be intervened between te and V2 in the V1 -te V2 predicates
(Martin 1975:510ff): -sae / -nado / -nanzo / -nanka / -nante / -gurai / -bakari / -dakewa / -made / -demo /
-wa / -mo
(26)
Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. 期末試験にもかかわらず，ケンは学校に鉛筆を持って { さえ/など/なんぞ/なんか/なん
て/は/も } いかなかった．
b. ケンは友達とのバーベキューパーティーに食材を何も買って { さえ/など/は } いかな
かった．
c. 町内会の清掃活動だというのに，ケンは道端のゴミを何も拾って { さえ/など/なんか/は/
も } いかなかった．
d. 日本に台風が近づいているにもかかわらず，ケンは外出前に天気予報を見て { さえ/な
ど/なんぞ/なんか/なんて/は/も } いかなかった．
10
e. 授業に遅れそうだというのに，ケンは学校に走って { さえ/など/なんぞ/なんか/なんて/
は/も } いかなかった．
(27)
5.2.2
After reanalysis: V-[ te-yuk ] (aspect)
a. *少年時代の思い出はケンの頭からなにひとつ消えて { さえ/など/なんぞ/なんか/なんて/
は/も } いかなかった．
b. *80 歳を過ぎてもケンの知的好奇心は失われて { さえ/など/なんぞ/なんか/なんて/は/も }
いかなかった．
c. *業界全体で不景気だというのに，その会社の株価は下がって { さえ/など/なんぞ/なんか/
なんて/は/も } いかなかった．
d. *人手不足にもかかわらず，社長は従業員を増やして { さえ/など/なんぞ/なんか/なんて/
は/も } いかなかった．
e. *材料費は上がっても，その商品の値段は上がって { さえ/など/なんぞ/なんか/なんて/は/
も } いかなかった．
Negation of V1
(28) Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. ケンは試験会場に鉛筆を持たないでいった．
b. ケンは友達とのバーベキューパーティーに食材を買わないでいった．
c. ケンは清掃活動の集合場所まで道端のゴミを拾わないでいった．
d. ケンは外出前に天気予報を見ないでいった．
e. ケンは学校に走らないでいった．
(29)
5.2.3
After reanalysis: V-[ te-yuk ] (aspect)
a. *少年時代の思い出はケンの頭からなにひとつ消えないでいった．
(cf. 少年時代の思い出はケンの頭からなにひとつ消えていかなかった．)
b. *80 歳を過ぎても，ケンの知的好奇心は失われないでいった．
(cf. 80 歳を過ぎても，ケンの知的好奇心は失われていかなかった．)
c. *業界全体で不景気だというのに，その会社の株価は下がらないでいった．
(cf. 業界全体で不景気だというのに，その会社の株価は下がっていかなかった．)
d. *人手不足にもかかわらず，社長は従業員を増やさないでいった．
(cf. 人手不足にもかかわらず，社長は従業員を増やしていかなかった．)
e. *材料費は上がっても，その商品の値段は上がらないでいった．
(cf. 材料費は上がっても，その商品の値段は上がっていかなかった．)
The o-V-ni nar Honorification
(30) Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. 山田先生が教室にそれを持っていった．
山田先生が教室にそれを { 持ってお行きになった / *お持ちになっていった }．
b. 山田先生が学生とのバーべ―キューパーティーに牛肉を買っていった．
山田先生が学生とのバーベキューパーティーに牛肉を { 買ってお行きになった /*お買い
になっていった }．
c. 山田先生が清掃活動の集合場所に道端のゴミを拾っていった．
山田先生が清掃活動の集合場所に道端のゴミを { 拾ってお行きになった /*お拾いになっ
ていった }．
d. 山田先生は外出前に天気予報を見ていった．
山田先生は天気予報を { 見てお行きになった / *ご覧になっていった }．
11
(31) After reanalysis: V-[ te-yuk ] (aspect)
a. 人手不足なので，多くの社長が従業員を増やしていった．
人手が足りないので，多くの社長が従業員を { *増やしてお行きになった / お増やしに
なっていった }．
b. 多くのお客様がこの商品を見ていった．
多くのお客様がこの商品を { *見てお行きになった / ご覧になっていった }．
c. 朝から多くのお客様がこの商品を買っていった．
朝から多くのお客様がこの商品を {*買ってお行きになった / お買いになっていった / お
買い上げになっていった }．
d. 会社の業績悪化に伴い，経営陣は従業員の数を減らしていった．
会社の業績悪化に伴い，経営陣は従業員の数を { *減らしてお行きになった / お減らし
になっていった }．
e. 山田先生は在職中多くの人材を育てていった．
山田先生は在職中多くの人材を { *育ててお行きになった / お育てになっていった }．
5.2.4
NPI-licensing
• That NPI-licensing across te phrases is possible means that the V1 -te V2 predicate behaves like
a single predicate (Nakatani 2013:12-13).
• To prove that the reanalyzed -te-yuk is morphologically more integral than the un-reanalyzed
-te-yuk, we expect the following contrast as to the NPI-licensing test:
– NPI-licensing across te is not allowed when V-te-yuk denotes movement after/while V.
– NPI-licensing across te is allowed when V-te-yuk denotes aspect.
(32) Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. ケンはバーべキューパーティーに食材を買っていった．
i. ケンはバーベキューパーティーに [ なにも 買わないで ] いった．
ii. ケンはバーベキューパーティーに [ なにも 買って ] いかなかった．
iii. *ケンは [ どこにも 食材を買わないで ] いった．
iv. ケンは [ どこにも 食材を買って ] いかなかった．
b. ケンは家に携帯電話を置いていった．
i. ケンは家に [ なにも 置かないで ] いった．
ii. ケンは家に [ なにも 置いて ] いかなかった．
iii. *ケンは [ どこにも 携帯電話を置かないで ] いった．
iv. ケンは [ どこにも 携帯電話を ] いかなかった．
c. ケンは集合場所に道端でゴミを拾っていった．
i. ケンは集合場所に [ 道端で 何も 拾わないで ] いった．
ii. ケンは集合場所に [ 道端で 何も 拾って ] いかなかった．
(33) After reanalysis: V-[ te-yuk ] (aspect)
a. 夜になって，気温が下がっていった．
i. *夜になっても気温が [ まったく 下がらないで ] いった．
ii. 夜になっても気温が [ まったく 下がって ] いかなかった．
b. 癌に侵され，その男の体はやせ細っていった．
i. *癌に侵されても，その男の体は [ まったく やせ細らないで ] いった．
ii. 癌に侵されても，その男の体は [ まったく やせ細って ] いかなかった．
c. 業績が回復し，その企業の株価は上がっていった．
12
i. *業績が回復しても，その企業の株価は [ まったく 上がらないで ] いった．
ii. 業績が回復しても，その企業の株価は [ まったく 上がって ] いかなかった．
d. 定年退職後，ケンは田舎暮らしを始めていった．
i. *定年退職後も，ケンは田舎暮らしを [ ちっとも 始めないで ] いった．
ii. 定年退職後も，ケンは田舎暮らしを [ ちっとも 始めて ] いかなかった．
5.2.5
Crossed Scrambling
• Expected results are as follows:
– For the un-reanalyzed -te-yuk (movement after/while V), crossed-scrambling is unavailable.
– For the reanalyzed -te-yuk (aspect), crossed-scrambling is available.
(34) Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. ケンは [ 食材を 買って ] 電車でバーベキューパーティーにいった．
i. ケンはバーベキューパーティーに [ 食材を 買って ] 電車でいった．
ii. *ケンは [ 食材を バーベキューパーティーに 買って ] 電車でいった．
iii. *ケンは電車で [ バーベキューパーティーに 食材を 買って ] いった．
iv. *ケンは電車で [ 食材を バーベキューパーティーに 買って ] いった．
v. *ケンは [ 食材を 電車で バーベキューパーティーに 買って ] いった．
b. ケンは [ 家に 携帯電話を 置いて ] 電車で試験会場にいった．
i. ケンは試験会場に [ 家に 携帯電話を 置いて ] 電車でいった．
ii. *ケンは [ 家に 携帯電話を 試験会場に 置いて ] 電車でいった．
iii. *ケンは電車で [ 試験会場に 家に 携帯電話を 置いて ] いった．
iv. ?ケンは電車で [ 家に 携帯電話を 置いて 試験会場に ] いった．
v. *ケンは [ 家に 携帯電話を 電車で 試験会場に 置いて ] いった．
c. ケンはゴミを拾って自転車で集合場所にいった．
i. ケンは集合場所に [ ゴミを 拾って ] 自転車でいった．
ii. *ケンは [ ゴミを 集合場所に 拾って ] 自転車でいった．
iii. *ケンは自転車で [ 集合場所に ゴミを 拾って ] いった．
iv. ?ケンは自転車で [ ゴミを 拾って 集合場所に ] いった．
v. *ケンは [ ゴミを 自転車で 集合場所に 拾って ] いった．
(35) After reanalysis: V-[ te-yuk ] (aspect)
a. 夜になって気温が 10 ℃に下がっていった．
i. 気温が 10 ℃に夜になって下がっていった．
b. 癌に侵されて彼の体重が 40 キロに減っていった．
i. 彼の体重が 40 キロに癌に侵されて減っていった．
5.2.6
Adjunct Modification of V2
(36) Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. ケンが学校に走って { 昨日/ゆっくり/のんびり/急いで/あわてて } 行った．
(37) After reanalysis: V-[ te-yuk ] (aspect)
a. 株価が上がって { *昨日/ *最近/ *どんどん/ *急激に/ *大幅に/ *過去最高に / *ゆっくり }
いった．
b. 株価が { 昨日/最近/どんどん/急激に/大幅に/過去最高に /ゆっくり } 上がっていった．
13
5.2.7
Phonological Contraction
(38) Before reanalysis: [ V-te ] -yuk (movement after/while V)
a. ケンが学校に走っていった．／ケンが学校に走ってった．
b. ケンは波消しブロックまで泳いでいった．／ケンは波消しブロックまで泳いでった．
c. ケンは机の上に財布を置いていった．／ケンは机の上に財布を置いてった．
(39) After reanalysis: V-[ te-yuk ] (aspect)
a. 株価が上がっていった．／株価が上がってった．
b. 気温が下がっていった．／気温が下がってった．
c. 彼の体重はみるみるうちに減っていった．／彼の体重はみるみるうちに減ってった．
5.3
Results of Observation
• The results of observation show that -te and -yuk is morphologically unified in the reanalyzed
V-te-yuk ‘V-Con-go’, which expresses aspectual meaning (See Tables 7 and 8).
Table 7: The results of observation as to the tests on complexity.
Before reanalysis: [V-te]-yuk
After reanalysis: V-[te-yuk]
Particle intervention
V1 negation
o-V-ni nar honorification
Ok
Ok
No
Ok
No
Ok
Table 8: The results of observation as to the tests on simplicity.
Before reanalysis: [V-te]-yuk
After reanalysis: V-[te-yuk]
NPI-licensing
Crossed-scrambling
V2 modification
Contraction
6
No
No
Ok
(Un)favorable
Ok
Ok
No
Favorable
Control Verb vs. Raising Verb
• In light of Kuno (1983), yuk ‘go’ behaves like a control verb in the un-reanalyzed V-te-yuk
‘V-Con-go’ whereas it behaves likes a raising verb in the reanalyzed -te conjunctive form.
• Control verbs vs. raising verbs (Kuno 1983):
(40)
Control verbs only allows the o-V1 -i V2 -ni-nar honorification:
a. 田中先生が手紙をお書き終えになった．
b. *田中先生が手紙をお書きになり終えた．
(Kuno 1983:10)
(41) Raising verbs only allows the o-V1 -i ni-nari V2 honorification:
a. ??大変です．団体のお客様がもうお着き始めになりましたよ．
b. 大変です．団体のお客様がもうお着きになり始めましたよ．
14
(Kuno 1983:12)
• Assuming from the contrast between (40) and (41), the un-reanalyzed V-te-yuk can behave like
a control verb because movement is a controllable event by the agent, while the reanalyzed
one can behave like a raising verb because it denotes an uncontrollable event. In fact, this
assumption proves to be true based on the contrast in honorification between (30) and (31).
References
Bybee, J. L., Perkins, R., & Pagliuca, W. 1994. The Evolution of Grammar: Tense, Aspect, and Modality in
the Languages of the World. Chicago: University of Chicago Press.
Cinque, G. 1999. Adverbs and Functional Heads: A Cross-Linguistic Perspective. Oxford: Oxford University Press.
Cinque, G. 2001. Restructuring and Functional Structure.
Linguistics, 11, 45–128.
University of Venice Working Papers in
Cinque, G. 2004. Restructuring and Functional Heads: The Cartography of Syntactic Structures, Vol. 4.
Oxford: Oxford University Press.
De Smet, H. 2012. Does innovation need reanalysis?. Leuven Working Papers in Linguistics, 1, 134–158.
Eckardt, R. 2007.
Grammaticalization
¡http://semanticsarchive.net/¿.
and
Semantic
Reanalysis.
.
Available
at
Gabelentz, G. v. d. 1891. Die Sprachwissenschaft. Ihre Aufgaben, Methoden, und bisherigen Ergebnisse.
Leipzig: Weigel.
Harris, A. C., & Campbell, L. 1995. Historical Syntax in Cross-Linguistics Perspective. Cambridge:
Cambridge University Press.
Haspelmath, M. 1998. Does grammaticalization need reanalysis?. Studies in Language, 22(2), 315–351.
Heine, B., Claudi, U., & Hunnemeyer,
F. 1991. Grammaticalization: A conceptual framework. Chicago:
¨
University of Chicago Press.
Heine, B., & Reh, M. 1984. Grammaticalization and Reanalysis in African Languages of the World. Hamburg:
Helmut Buske.
Hopper, P. J. 1991. On some principles of grammaticization. In Traugott, E. C., & Heine, B. (Eds.),
Approaches to Grammaticalization, Vol. 1, pp. 17–35. Amsterdam/Philadelphia: John Benjamins.
Hopper, P. J., & Traugott, E. C. 2003. Grammaticalization (2 edition). Cambridge: Cambridge University
Press.
Kageyama, T. 1993. Bunpoo to Gokeisei. Tokyo: Hituzi Syobo.
Kuno, S. 1983. Sin Nihon Bunpoo Kenkyuu. Tokyo: Taishukan Shoten.
Langacker, R. W. 1977. Syntactic reanalysis. In Li, C. N. (Ed.), Mechanisms of Syntactic Change, pp.
57–139. Austin: University of Texas Press.
Langacker, R. W. 1990. Subjectification. Cognitive Linguistics, 1, 5–38.
Lichtenberk, F. 1991. Semantic change and heterosemy in grammaticalization. Language, 67, 475–509.
Longobardi, G. 2001. Formal Syntax, Diachronic Minimalism, and Etymology: The History of French
chez. Linguistic Inquiry, 32(2), 275–302.
15
Martin, S. E. 1975. A reference grammar of Japanese. Connecticut: Yale University Press.
Matsumoto, Y. 1996. Complex Predicates in Japanese: A Syntactic and Semantic Study of the Notion ‘Word’.
Stanford, CA: CSLI.
McCawley, J. D., & Momoi, K. 1986. The Constituent Structure of -te Complements. Papers in Japanese
Linguistics, 11, 1–60.
Meillet, A. 1912. L’´evolution des formes grammaticales. Scientia (Revista di Scienza), 12(26, 6).
Reprinted in Meillet, Antoine (1958) Linguistique historique et linguistique g´en´erale (Collection
Linguistique publi´ee par la Soci´et´e de Linguistique de Paris 8), Paris: Champion, pp. 130-148.
Miyagawa, S. 1987. Restructuring in Japanese.
Linguistics, pp. 273–300. Dordrecht: Foris.
In Imai, T., & Saito, M. (Eds.), Issues in Japanese
Nakatani, K. 2006. Processing Complexity of Complex Predicates: A Case Study in Japanese. Linguistic Inquiry, 37, 625–647.
Nakatani, K. 2013. Predicate concatenation: A study of the V-te-V predicate in Japanese. Kurosio Publishers,
Tokyo.
Nishigauchi, T. 2009. The Awareness Condition and the POV Projections. TALKS: Theoretical and
Applied Linguistics at Kobe Shoin, 12, 37–49.
Nishigauchi, T. 2014. Reflexive binding: awareness and empathy from a syntactic point of view.
Journal of East Asian Linguistics, 1–50.
Roberts, I. 2010. Grammaticalization, the clausal hierarchy and semantic bleaching. In Traugott,
E. C., & Trousdale, G. (Eds.), Gradience, Gradualness and Grammaticalization, pp. 45–73. Amsterdam/Philadelphia: John Benjamins.
Roberts, I., & Roussou, A. 2003. Syntactic Change: A Minimalist Approach to Grammaticalization.
Cambridge: Cambridge University Press.
Smith, A. 2006. The universal tendency for renewal among grammatical expressions for anterior and
related aspect. Journal of Universal Language, 7, 139–160.
Sweetser, E. E. 1988. Grammaticalization and semantic bleaching. In Axmaker, S., Jaisser, A., &
Singmaster, H. (Eds.), Berkeley Linguistics Society 14: General Session and Parasession on Grammaticalization, pp. 389–405. Berkeley: Berkeley Linguistics Society.
von Fintel, K. 1995. The formal semantics of grammaticalization. NELS, 25, 175–189.
16
© Copyright 2025 ExpyDoc
