Presupposition Accommodation by Discourse Initial Evidential Marker Mo

1
Presupposition Accommodation by Discourse Initial
Evidential Marker Mo
International Conference on Revisiting Japanese
Modality
June 25, 2006 University of London, SOAS

• Sumiyo Nishiguchi
• Stony Brook University
• snishigu_at_ic.sunysb.edu

2
Abstract

• 1. A focus marker mo also/evencan associate
  with the whole proposition (Numata 2000) and
  appear discourse initially.
• 2. Mo triggers a set of stative propositions
  similar to the asserted one as presuppositions
  (cf. Shudo 2002).
• The hearer accommodates the informative
  presuppositions.

3

• 3. This kind of mo is an evidential marker.
• 4. Mo is a modal morpheme of evidentiality as
  well as speaker's sentiments.
• 5. Mo takes unaccusative predicates.

4

• 6. Mo is a determiner which takes presupposition
  in the restrictor and the unaccusative predicates
  in the nuclear scope.
• 7. Mo-p updates the information state.
• 8. The illocutionary force of mo-p is not only to
  report (cf. Faller 2002) but to prompt actions.

5

• 1. Mo

6
Mo1

• Mo also/even is a particle attached to noun
  phrases in Japanese.
• Mo also/even is a focus marker or a
  quantifier-like-element (Kuroda1969).
• Mo1 also
• (1) Ken-mo ki-ta.
• Ken-also come-PAST
• Ken came, too'

7
Mo2

• Mo obtains the meaning of even' when the NP is
  focused (Watanabe 2004).
• Mo2 even
• (2) KenF-mo ki-ta.
• Ken-also come-PAST
• Ken came, too'

8

• The even' mo forms NPIs with indeterminates
  (wh-words) (Kuroda 1965 Watanabe 2004 cf.
  Lahiri 1998)
• (3) Dare-mo ko-nai.
• who-even come-NEG
• Nobody comes'

9

• Mo with minimum quantity serves as a minimizer
  (Kato 1985)
• (4) Hito-ri-mo ko-nai.
• 1-CL-even come-NEG
• Nobody comes'

10

• Japanese grammarians have considered mo
  polysemous (Sadanobu 1997)
• i) also ii) even and, iii) attenuation
• (Numata 1986) or admiration (Teramura 1991).

11
Mo3

• The meaning of the third mo is distinct from the
  other two, also and even.
• Mo can associate with the whole proposition
  (Numata 1986, Numata 2000).

12
Wide Scope and Sentence Focus

• (5) Yo-mo hukete-ki-ta.
• night-also pass-come-PAST
• Mo neru-to shi-yo.
• already sleep-COMP do-will
• Its late. I will go to bed'
• (5) LF mo yo-ltmogt huke-ta

13
No Antecedent

• This kind of mo does not need any antecedent,
  such as x has become late, and the night also
  grew late.''

14
Sentence-Focus Structure (Lambrecht 2002)

• (6) a. Sentence My CAR broke down.
• b. Presupposition ___
• c. Assertion speakers car broke down
• d. Focus domain speakers car broke
  down
• e. Focus S
• (Lambrecht 2002 233)

15
Moreover, Discourse Initial

• This sentence focus mo is used discourse
  initially
• (7) Haru-mo takenawa-ni nari-mashi-ta
• spring-also peak-GOAL become-HON-PAST
• The spring has reached its peak'
• (Numata2000 172)

16
What is the meaning of mo?

• Mo3 has its distinct meaning since replacing mo3
  with ga (nominative case) or wa (topic marker)
  changes the meaning.
• (8)a. Haru-mo takenawa-ni nari-mashi-ta
• spring-also peak-GOAL become-HON-PAST
• b. Haru-ga takenawa-ni nari-mashi-ta.
• spring-NOM peak-LOC become-HON-PAST
• C. Haru-wa takenawa-ni nari-mashi-ta.
• spring-TOP peak-LOC become-HON-PAST
• The spring has reached its peak'

17
Observations

• Mo triggers presuppositions (section 2)
• Mo expresses sentimentality (section 3)

18

• 2. Presupposition Accommodation by Mo

19

• Numata argues that (7) either
• i) evokes other events relevant
• to change of seasons or,
• ii) pretends the existence of
• antecedents in order to give rise to attenuating
  effect
• (Numata1986, Numata2000)

20
What does mo presuppose?

• The speaker presupposes prior situations
• e.g., cherry blossoms are blooming, the leaves
  became green, and the weather became warmer.

21

• (9) Yo-mo fukete-ki-ta. Mo neru-to shi-yo.
  night-also pass-ASP-PAST
  already sleep-COMP do-will
• Its late. I will go to bed'

22

• (10) Soto-mo hiete-ki-ta.
• outside-also cold-ASP-PAST
• It has become cold outside'

23

• (11) Ko-no saifu-mo furuku-nat-ta.
• this-GEN wallet-also old-become-PAST
• This wallet has become old

24

• (12) Tabi-mo owari-ni chikazuite-ki-ta.
• trip-also end-to approach-ASP-PAST
• The trip is nearing the end
• (13) Omae-mo aho-ya-na.
• you-also silly-be-EXC
• You are silly, I should say'

25

• (14) Yo-mo sue-da.
• world-also finale-be
• This is the end of the world

26
Mo is a presupposition trigger

• (15) Mo Yo-ga fuke-ta its late It is dark,
  It is past midnight, The neighbors turned off
  their lights
• (16) Mo Haru-ga takenawa-ni natta spring
  reached its peak' Cherry blossoms are
  blooming, It became warmer, The daylight has
  become longer,...

27

• (17) Mo ko-no saifu-ga huruku-nat-ta
• It has been long since I got this, the wallet
  looks worn and torn,
• (18) Mo tabi-ga owari-ni chikazui-ta
• the train is approaching the destination,

28
Presuppositions of Mo

• When uttering mo-p discourse initially
• - Speaker acts as if presupposed preceding events
  are part of the common ground (cf. Stalnaker
  1973, 1974)
• - Speaker presupposes that the auditor will be
  able to infer that it is presupposed.

29
Presupposition Accommodation

• Lewis (1979)
• If a time t something is said that requires
  presupposition P to be acceptable, and if P is
  not presupposed just before t, ceteris paribus
  and within certain limits- presupposition P comes
  into existence at t.

30

• Mo-p triggers presupposition that are
  propositions similar to p.
• Eg.,Haru-mo takenawa-ni narimasita Spring has
  reached its peak
• presupposes
• Cherry blossoms are blooming, it became
  warmer,

31

• What does mo quantify over?
• - Mo quantifies over unspecified events.
• The presuppositions might differ between
  the hearer and the speaker.
• Japanese cherry blossoms Americans
  longer daytime
• - Mo quantifies over evidences.
• - Mo quantifies over the near and similar
  possible worlds.

32
Comparative similarity (Lewis 1973)

• SIMw is a function from propositions to
  propositions which maps each p to the set of
  p-worlds similar to w.
• (19) SIMw pw?p wis similar to w no less
  than any other world in p
• (20) ?WW, for w?W,
• u w v iff u is more similar to or close to w
  than v
• (21) mo-p(w)1 iff ?u?p, ?v?W s.t. u w v
  q(v)1

33

• The addressee supplements the presupposed
  propositions into his knowledge (presupposition
  accommodation)

34

• The hearer has not been (at least fully) aware of
  the presupposed events until the speaker utters
  mo-p.
• When hearing p, the addressee recognizes the
  evidence that the bag is torn or the spring has
  reached its peak.

35
Informative Presuppositions (Stalnaker 1988)

• Normally, presuppositions are not informative.
• However, the presuppositions of mo-p are
  informative.
• Mo-p updates the common ground. The
  presuppositions of mo-p are added to the hearers
  knowledge.

36
Summary on Section 2 Presupposition Accommodation

• Mo triggers unspecified presuppositions similar
  to the asserted proposition
• The hearer infers the likely resuppositions.

37

• 3. Mo is an evidential marker

38
Mo is an Evidential Marker

• Building on Numata's insight, I claim that this
  kind of mo is an evidential marker.
• (7) is an utterance when speaker and
• hearer have perceived evidence that spring has
  reached its peak. visual

39

• (7) Haru-mo takenawa-ni nari-mashi-ta
• spring-also peak-GOAL become-HON-PAST
• The spring has reached its peak'
• (Numata2000 172)

40
What is evidentiality?

• Evidentiality the indication of speakers source
  of information
• A true evidential encodes a type of information
• Scales
• i)Visualgtauditorygtother sensorygtinference from
  resultsgtreasoning
• ii) Directgtsecondhandgtthirdhandgthearsay/
• folklore
• (Faller 2002)

41

• (9) Yo-mo fukete-ki-ta. Mo neru-to shi-yo.
  night-also pass-ASP-PAST already sleep-COMP
  do-will
• Its late. I will go to bed'

42

• (9) would be uttered after the speaker had a look
  at a clock. direct visual evidence
• (9) and (7) are reports based on direct evidence,
  aimed to draw hearer's attention to time or a
  season.

43

• (11) Ko-no saifu-mo huruku-nat-ta.
• this-GEN wallet-also old-become-PAST
• This wallet has become old
• visual direct evidence

44

• (12) Tabi-mo owari-ni chikazuite-ki-ta.
• trip-mo end-to approach-ASP-PAST
• The trip is nearing the end
• direct nonvisual evidence
• (13) Omae-mo aho-ya-na.
• you-mo silly-be-EXC
• You are silly, I should say
• direct inferential evidence

45

• Then, mo is an evidential marker of
• direct, mostly visual evidences.

46
Evidential markers in Japanese

• Aoki (1986)
• so hearsay
• gar hearsay and inferential form
• rashi circumstantial evidence or gathered
  through sources other than ones own senses

47
Epistemic must

• Speaker believes must p based on presently
  available evidence
• (cf. Coates 1983, Woisetschlaeger 1985, on
  epistemic must)
• (20) believe(speaker, must p)

48
Conversational Backgrounds

• (21) Modal base f (in view of the direct
  evidences) must p

49

• 4. Sentimentality

50
Sentimental/Bouletic Modality

• Not only being evidential, mo is an expression of
  sentimental modality.
• While (7) demonstrates speaker's uplifted
  sentiments, most of the evidential mo sentences
  express speaker's negative emotion.

51

• (11) Ko-no saifu-mo huruku-nat-ta.
• this-GEN wallet-also old-become-PAST
• This wallet has become old
• sentiments
• (9) Soto-mo hiete-ki-ta.
• outside-mo cold-ASP-PAST
• It has become cold outside
• negative

52

• (12) Tabi-mo owari-ni chikazuite-ki-ta.
• trip-mo end-to approach-ASP-PAST
• The trip is nearing the end
• sentiments, negative
• (13) Omae-mo aho-ya-na.
• you-mo silly-be-EXC
• You are silly, I should say
• sentiments

53
Ordering source (cf. Kratzer1991)

• (22)Ordering source g (in view of what I want)
  For all u, w, w?W, for any g(u)?P(W)
• wg(u)w iff
• p p?g(u) w?p ?pp?g(u) and w ?p

54

• (23) mo-f(w) 1 iff
• for all u?ng(w) there is a v ?ng(w) such that
  vg(w)u and for all z ?ng(w) if zg(w)v, then
  z?f.

55

• (24)
• f
• g ordering source bouletic
• MODAL f modal base evidential
  modal
• (cf. Kratzer 1991 von Fintel and Iatridou 2005)

56

• 5. Unaccusativity

57
Unaccusativity

• All predicates with wide scope mo are
  unaccusative.
• fukeru become late
• takenawa-ni naru reach its peak
• furuku naru become old
• owari-ni chikazuku near the end
• hiete kuru become cold
• aho-da be silly

58
No passive morpheme

• (25) Yo-ni fuke-rare-ta.
• night-DAT grow late-PASS-PAST
• (26) Kono-kaban-ni furuku-nar-are-ta.
• this-bag-DAT old-become-PASS-PAST
• (27) Tabi-ni owari-ni chikazuk-are-ta.
• trip-DAT end-LOC near-PASS-PAST

59

• 6. Generalized Quantifier

60
Restrictive Quantification

• Bouletic and evidential modal is the lexical
  meaning of mo.
• Wide scope mo functions as a determiner taking
  implicit presupposition in the
• restrictor and overt stative or resultative
  unaccusative predicates in the nuclear scope.
• (cf. Kratzer 1991 Berman 1991 von Fintel 1994)

61

• (28) TP
• Mo P VP
• Mo (MODALbouletic,evidential) presupposition

62

• (29) TP
• moP TP
• moP T
• mo presupposition yo ltmogt
• VP T
• ltyogt huke ta

63

• Mo takes the propositions similar to p in the
  restrictor.
• Mo takes the set of possible worlds in which the
  presuppositions are true as a restrictor.
• (30)Mo (?w. SIMw(p))(?w. p(w))
• 1 if there is w such that SIMw(p)1
• else 0.

64
Kuroda (1986)

• Mo is in COMP in Deep Structure
• (31) DS Hanako-ga odor-ta shi mo
• Hanako-NOM dance-PAST and also
• Masao-ga odor-ta mo
• Masao-NOM dance-PAST also
• SS Hanako-mo odot-ta-shi
• Hanako-also dance-PAST and
• Masao-mo odot-ta.
• Masao-also dance-also do-PAST
• Hanako danced and Masao danced, too

65

• 7. CCP
• and
• Information Updates

66
Context Change Potential (development from
Heim 1992)

• (32) cmo-p(w,g) in c p and
  Believe(sh,must(Know(sh,p))) is true at (w,g)
• if there is q which is similar to, but not equal
  to p which is true in c otherwise, undefined,
• where s is the speaker and h the hearer, g is
  the assignment function and for MBelieve or
  Know, M(ab,p) is true at (w,g) iff for all u,
  w(RM,a U RM,b) u implies that p is true at
  (u,g)
• ((Rx U Ry) is the union of Rx and Ry, and (Rx)
  is the reflexive-transitive closure of Rx

67
Information Update

• Utterances with mo draws the hearer's attention
  to the fact which the hearer had not been so
  aware of, e.g., its late, or the spring reached
  its peak,
• On the other hand, the sentence without mo is a
  statement of the fact and does not convey any new
  information. p resides within the common
  knowledge between the speaker and the hearer (cf.
  Gronedijk and Stokhof 1991 Veltman 1996).

68

• Mo-p adds I believe that we must know that p
• (33) BELIEVE MUST KNOW (speaker and hearer, p)

69

• 8. Illocutionary Force

70
Speech Act Function of Mo

• Faller 2006 The illocutionary force of Cusco
  Quechua reportative evidentials is report
• The speech act functions of mo is to draw
  attention of the hearer to the evidenced fact
  that becomes a reason for the following sentence
  and prompt actions.

71

• (34) Kane-mo morae-nakat-ta-shi kaer-o.
• money-also receive-NEG-PAST-so return-will
• Having been given no money, let us go home
• (35) Yo-mo huke-ta-kara ohiraki-ni shi-yo.
• night-also pass-PAST-so finish-GOAL do-lets
• Its late. Lets wrap up a party

72

• (33) REPORT(?1 ) ? BECAUSE(?1, ?2)
• In the framework of SDRT
• (Asher and Lascarides 2003)

73
Conclusion

• Mo is an evidence invoking marker
• The hearer evokes evidence for p
• and believes that p is necessarily true
• Mo sentences are subjective.
• Speaker feelings, often negative ones, are
  expressed.

74

• Speaker Modality incorporated into mo is a
  determiner which takes implicit presuppositions
  in the restrictor and the unaccusative VPs in the
  nuclear scope

75

• Wide scope taking mo triggers similar
  propositions to p as presuppositions
• The presuppositions are informative.
• The hearer infers the missing presuppositions
  from extra-linguistic contexts, thus,
  accommodates presuppositions

76

• Mo-p updates common ground
• The illocutionary force of mo-p is to give
  reasons for action.