An illocutionary account of reportative evidentials in...

An illocutionary account ofreportative evidentials in

imperatives

Scott AnderBois

Brown UniversitySALT 27 – University of Maryland

May 14, 2017

§1 Introduction

Reportatives in imperatives

I In declarative sentences, evidentials encode the informationsource for a given claim.

I In interrogatives, languages often allow only a subset ofevidentials. (Aikhenvald, 2004)

I Nonetheless, the range of interpretations of evidentials ininterrogatives across languages have given vital analyticalinsights (e.g. Murray (2011), Murray (2010), Lim (2010))

I In imperatives, most languages disallow all evidentials, withonly reportative evidentials widely attested:

(1) Kumuhatake.Agt.Trig

kayou.Dir

daw

RepngIndir

tinapay.bread

‘Take some bread (she says/they say)!’ [Tagalog]

§1 Reportatives in declaratives

Reportatives lack an overt reporter

I Reportative evidentials express information which issecond-hand (or third-hand).

I Unlike verbs of saying and quotative particles, however, theydo not allow for the reporter to be made explicit (e.g. as anargument or via agreement):

(2) Scenario: I was talking to my friend Luis earlier about the Xtaabay (amythical woman who seduces and attacks drunk men in the jungle)and now tell you:Chowaklong

bin

RepuA3

tso’ots-elhair-Rel

uA3

poolhead

leDef

ixtaabay=o’.Xtaabay=Distal

‘I was told (by Luis) that the Xtaabay’s hair is long.’ [Yucatec Maya]





(2) Scenario: I was talking to my friend Luis earlier about the Xtaabay (amythical woman who seduces and attacks drunk men in the jungle)and now tell you:*Chowaklong

bin

Rep{leti’/Luis}him/Luis

uA3


uA3

poolhead

leDef

ixtaabay=o’.Xtaabay=DistalIntended *‘I was told by him/Luis that the Xtaabay’s hair is long.’[Yucatec Maya]





(2) Scenario: I was talking to my friend Luis earlier about the Xtaabay (amythical woman who seduces and attacks drunk men in the jungle)and now tell you:*Chowaklong

u

A3bin

RepuA3


uA3

poolhead

leDef

ixtaabay=o’.Xtaabay=Distal

Intended *‘I was told by him/her that the Xtaabay’s hair is long.’[Yucatec Maya]





(2) Scenario: I was talking to my friend Luis earlier about the Xtaabay (amythical woman who seduces and attacks drunk men in the jungle)and now tell you:*Luis=e’

Luis=Top

chowaklong

bin

RepuA3


uA3

poolhead

leDef

ixtaabay=o’.Xtaabay=DistalIntended *‘As for Luis, I was told by him that the Xtaabay’s hair islong.’ [Yucatec Maya]


Reportatives have apparently anaphoric uses

I Rep don’t license ‘reporter’ arguments, but they allow for(apparently) anaphoric uses, (3) (e.g. Murray (2010), Schwager

(2010))

I Rep do not require a salient reporter, allowing for ‘hearsay’uses with the reporter unspecified, (4).

(3) Scenario: We are inside a windowless building. I just got o↵ thephone with my friend Mary who is outside and tell you:Umuulanrain.Impf

daw.Rep

‘It’s raining outside, Mary says.’ [Tagalog]

(4) Scenario: We are talking our opinions about a local politician.Matapathonest

daw

Repsiya.Dir.3sg

‘He’s honest, I heard.’ (alt. ‘He’s honest, they say.’) [Tagalog]


Scope proposition deniable in declaratives

I Rep are typically used as though the speaker believes thescope proposition possibly or likely true (e.g. Faller (2007))

I Rep allow the scope proposition to be explicitly denied withsu�cient perspective shifting, (5) (e.g. AnderBois (2014))

(5) a. Scenario: We are talking our opinions about a local politician.Matapathonest

daw

Repsiya,Dir.3sg

perobut

hindiNeg

namanContr

iyonthat

totoo.true

‘He’s honest, they say, but it’s not really true.’ [Tagalog]

b. Scenario: We are talking about a recent exam.

i. Ma’Neg

bin

Rept-inPfv-A1

maans-(aj)-epass-Status-Def

eeksaamen=o’exam=Distal

‘I didn’t pass the exam reportedly, . . . ’

ii. . . . ba’ale’. . . but

t-uPrep-A3

jaaj-il=e’true-Rel=Top

t-inPfv-A1

maans-aj.pass-Status

‘. . . but actually I passed.’ [Yucatec Maya]

§1 Imperatives

Imperatives defined

I Imperatives are a major syntactic sentence type acrosslanguages

I Although imperatives are often most associated with directivespeech acts like commands, . . .

I . . . they are consistently polyfunctional cross-linguistically:(see Aikhenvald (2010), Condoravdi & Lauer (2012) for recent overviews)

(6) Imperative speech acts:

a. Stand at attention! Command

b. Please, don’t be rain! Wish

c. Have a cookie. Offer

d. Take the train that leaves in 10 minutes. Advice

§1 Imperatives

Imperatives defined

I Imperatives are a major syntactic sentence type acrosslanguages

I Although imperatives are often most associated with directivespeech acts like commands, . . .

I . . . they are consistently polyfunctional cross-linguistically:(see Aikhenvald (2010), Condoravdi & Lauer (2012) for recent overviews)

(6) Imperative speech acts:

a. Stand at attention! Command

b. Please, don’t be rain! Wish

c. Have a cookie. Offer

d. Take the train that leaves in 10 minutes. Advice

§1 Imperatives

Imperatives w/ reportatives

I ImpRep like (7) little discussed in previous literature:

(7) Context: A family member o↵ers advice to a teenagerMaligobathe.Imper

kaDir.2sg

nanow

raw

Rep‘Take a bath (s/he says).’ [Tagalog]

I Two informal semantic characterizations of ImpRep:

Imperative by proxy: ImpRep ‘not just ‘hearsay’ — it implies acommand to do something on someone else’s order’(Aikhenvald, 2010, p. 138)

Neutral report: ImpRep ‘not an imperative on behalf of a thirdparty, but rather an entirely neutral report of animperative’ (Schwager, 2010, p. 8) (see also Thomas (2014),

Korotkova (2015), Korotkova (2016))

§1 Imperatives

Imperatives w/ reportatives

I ImpRep like (7) little discussed in previous literature:

(7) Context: A family member o↵ers advice to a teenagerMaligobathe.Imper

kaDir.2sg

nanow

raw

Rep‘Take a bath (s/he says).’ [Tagalog]

I Two slightly more formal reworkings of thesecharacterizations:

Imperative by proxy: ImpRep contribute the same primarymeaning/CCP as other imperatives.

Neutral report: ImpRep contribute informational updates withreportative or quotative contents.

§1 This talk

I Today: Argue that ImpRep – at least in Tagalog and YucatecMaya – best fit the ‘imperative by proxy’ characterization, apattern which calls for an illocutionary account.

Road map:

§2 examines the illocutionary e↵ects of ImpRep indiscourse including the range of responses they elicit;

§3 looks at two di↵erent types of ‘weakening’ claimedto happen in ImpRep;

§4 develops an illocutionary analysis of ImpRep

extending AnderBois (2014)’s scoreboard semanticsfor reportatives in declaratives;

§5 concludes with implications for the typology ofevidentials.

§1 Language background

I Naturally occurring examples and context-relative felicityjudgment data from two languages:

Yucatec Maya (Mayan, Mexico)

I Reportative bin part of asmall set of clitics whichoccur at the end of anyprosodic word

Tagalog (Austronesian, Philippines)

I Reportative daw/raw partof a set of 18 secondposition discourse particles(Schachter & Otanes (1972))

§2 Illocutionary force of ImpRep

ImpRep are polyfunctional

I As the name ‘imperative by proxy’ suggests, ImpRep can beused to issue commands to the addressee:

(8) Order Scenario: Our mother has told me to make sure that myyounger sibling eats their dinner. After talking to her, I tell my sibling:Uk’drink.Imper

bin

Repayour

wo’ochmeal

k’eyem=o’pozole-Distal

‘Drink your pozole (she orders)!’ [Yucatec Maya]

I While Aikhenvald (2010) focuses exclusively on commanduses, recalling the polyfunctionality above, we find thatImpRep can be used in other imperative speech acts:





bin

Repayour

wo’ochmeal




(9) Advice Scenario: Your stomach hurts and I am on the phone withour Mom asking for advice. I tell you:Uk’drink.Imper

bin

Repayour

wo’ochmeal


‘Drink your pozole (she says)!’ [Yucatec Maya]





bin

Repayour

wo’ochmeal




(10) O↵er Scenario: A child’s mother has told them they should o↵ercake to guests. The child says to a guest:Jaanteat.Imper

bin

RepleDef

paastel=a’cake=Distal

‘Eat this cake (Mom told me to o↵er it)! [Yucatec Maya]





bin

Repayour

wo’ochmeal




(11) Well-wish scenario: My friend wants me to tell you to take care. Isay to you:Kalaant-a-bajcare.for-A2-Refl

bin!Rep

‘Take care (s/he says)!’ [Yucatec Maya]





bin

Repayour

wo’ochmeal




(12) Optative: A friend told me to wish you well. I say:KaaIrr

xi’ikgo.Subj

bin

ReptechDat.B2sg

utsgood

‘May you be well (my friend wishes)!’ [Yucatec Maya]





bin

Repayour

wo’ochmeal




(13) Exhortative: My friend told us Ko’ox janal! ‘Let’s eat!’. You didn’thear what was said and so I repeat for you:Ko’oxcome.Hort

bin

Repjanal.eat

‘Let’s eat (she says).’ [Yucatec Maya]





bin

Repayour

wo’ochmeal




(14) Negative imperative: Juan says he is going to kill one of his father’spigs. I am on the phone with Juan’s dad and say:Ma’Neg

bin

RepaA2

kıinsikkill

(leDef

k’eek’en=o’)!pig=Distal

‘Don’t kill the pig (he says)!’ [Yucatec Maya]

§2 Infelicitous for information-seeking

Not interchangeable with declaratives

I In principle, ImpRep contain information about what thereporter wants to have happen (and/or what they said theywant).

I If ImpRep are neutral reports, we should expect them to befelicitous in contexts where the addressee seeks thisinformation. However:

(15) #Information-seeking scenario: You talked to Mom, what did youhear?#Xeengo.Imp

bin

RepaA2

maanbuy.Subj

bu’ulbean

‘Go buy beans (I heard).’ [Yucatec Maya]

I Analogous declaratives with verba dicendi, deontic modals, ordesideratives plus Rep are fine in this case.





(16) #Descriptive bouletic modal scenario: You talked to Mom, whatdoes Mom want?#Xeengo.Imp

bin

RepaA2

maanbuy.Subj

bu’ulbean







(16) #Descriptive bouletic modal scenario: You talked to Mom, whatdoes Mom want?#Xeengo.Imp

bin

RepaA2

maanbuy.Subj

bu’ulbean



§2 Responses to ImpRep

Possible overt responses pattern with imperatives

I While declarative sentences do not oblige an overt response,they allow responses such as ‘yes’ and ‘no’ (e.g. Farkas & Bruce

(2010) et seq.)

I If neutral reports, ImpRep ought to allow for such responses.

I However, we find that such responses are infelicitous, whereasresponses possible for ordinary imperatives are acceptable:

(17) a. A: Uk’drink.Imper

bin

Repayour

wo’ochmeal



b. B: Ma’alobokay

tuun.then

// #Jaaj.true

‘Okay then.’ // #‘True.’

§2 Responses to ImpRep

Possible overt responses pattern with imperatives

I While declarative sentences do not oblige an overt response,they allow responses such as ‘yes’ and ‘no’ (e.g. Farkas & Bruce

(2010) et seq.)

I If neutral reports, ImpRep ought to allow for such responses.

I However, we find that such responses are infelicitous, whereasresponses possible for ordinary imperatives are acceptable:

(17) a. A: Uk’drink.Imper

bin

Repayour

wo’ochmeal



b. B: Ma’,No

(minNeg.1sg

jantik).eat

// MixNeg

taan.Prog

// #Ma’Neg

jaaj=i’.true=Neg.Cl‘No, (I won’t).’ // ‘I won’t.’ // #‘It’s not true.’

§2 Cross-linguistic uniformity?

Tagalog daw

I Schwager (2010) claims Tagalog ImpRep are neutral reportsand o↵ers the following judgment as support:

(18) a. K<um>aineat<AV>

(ka)you

daw.Rep

“e.g. ‘Mommy/They/. . . said that you should eat.’ ” [Tagalog]

b. HindiNeg

totoo!true

‘That’s not true!’

§2 Cross-linguistic uniformity?

Tagalog daw

I Schwager (2010) claims Tagalog ImpRep are neutral reportsand o↵ers the following judgment as support:

I However, in my own fieldwork, speakers find this responsemarginal at best, and show the same pattern as in YM:

(19) Order Context: Our mother has told me to make sure that myyounger sibling eats their bread. I tell my sibling:

a. Kainineat.PT

moIndir.2sg

daw

RepangDir

tinapaybread

moyour

‘Eat you bread (she orders)!’ [Tagalog]

b. #HindiNeg

(totoo)true

// Ayawnot.want

koIndir.1sg

// #Ooyes

// #Totootrue

iyanthat

// Sigeokay

(na)now

(nga)indeed

# ‘It’s not true.’ // ‘I don’t want to.’ // #‘Yes’ // # ‘That’strue.’ // Okay. [Tagalog]

§3 ‘Softening’ of imperatives

Aikhenvald 2004/2010’s claims

I Aikhenvald claims that reportatives may ‘attenuate’ or ‘soften’commands, implying that this possibility varies by languages:

(20) marna-luspinifex-3pl

ma-ntaget-Imper

nganta?Rep

‘Pick up the spinifex, won’t you?’[Warlpiri]

(Laughren, 1982, p. 138)

(21) Arrantherre2plS

kwele

Repntert-irr-;-aye!quiet-Inch-Imper-Emph

‘You mob are supposed to be quiet.’[Mparntwe Arrernte]

(Wilkins, 1989, p. 393)

I N.B. it’s not clear how/if this claim extends to other speechacts like o↵ers and advice.

§3 ‘Strength’ and deniability

Context modulates ‘strength’ indirectly

I For YM bin and Tagalog daw, however, we find no evidenceof such a specific relationship.

I Rather, the perceived ‘strength’ is modulated by the social orrational authority of the reporter (cf. Hamblin (1987), Kaufmann

(2012) on ordinary imperatives):

I Indefinite/generic reporters often lead to ‘weakening’, but thisis not a property of the reportative per se.

(22) Strong Social Authority Order Scenario: Our mother has told meto make sure that my younger sibling eats their dinner. After talkingto her, I tell my sibling:Uk’drink.Imper

bin

Repayour

wo’ochmeal









(23) Weak Social Authority Order Scenario: Our younger brother hastold me to make sure my uncle eats his dinner. After talking to mybrother, I turn to my uncle and say:Uk’drink.Imper

bin

Repayour

wo’ochmeal


‘Drink your pozole (he orders)!’ [Yucatec Maya]







(24) Unspecified Reporter Order Scenario: We are sitting at the tableand you aren’t having your pozole. I say to you:Uk’drink.Imper

bin

Repayour

wo’ochmeal


‘Drink your pozole (they say)!’ [Yucatec Maya]


Deniability

I Related to the preceding observations, Thomas (2014) arguesfor a ‘neutral report’ approach in part by citing the followingexplicit denial:

(25) E-me’e2.Imper-give

je

Repchevyme

peto

ka’ygua,mate,

va’eribut

nd-a-ipota-i.Neg-A1-want-Neg

‘Give me the mate, I heard, but I don’t want it.’ [Mbya]

I Similar such denials are somewhat possible in Yucatec Mayaand Tagalog:

(26) Jaanteat.Imper

bin

RepaA2

wo’och=o’,meal=Distal

ba’ale’but

ten=e’I=Top

ma’Neg

inA1

k’aatwish

kaaIrr

aA2

jaant=i’.eat=Neg.Cl

‘Eat your meal (he says), but I don’t want you to eat it.’ [Yucatec

Maya]


Deniability



je

Repchevyme

peto

ka’ygua,mate,

va’eribut



I Similar such denials are somewhat possible in Yucatec Mayaand Tagalog:

(27) ?Tumakborun.Imper

kaDir.2sg

daw

Reparaw-araw,daily

perobut

ayawnot.want

kita-ngDir2sg.Indir1sg-Lnk

tumakbo.run

‘Run daily (they say), but I don’t want you to.’ [Tagalog]


Deniability



je

Repchevyme

peto

ka’ygua,mate,

va’eribut



I . . . and more marginally with conflicting imperatives:

(28) ?Jaanteat.Imper

bin

Repverduras,vegetables

ba’ale’but

(t-uPrep-A3

jaaj-il=e’)true-Rel=Top

ma’Neg

aA2

jantik.eat

‘They say to eat your vegetables, but (really) don’t.’ [Yucatec Maya]


Deniability



je

Repchevyme

peto

ka’ygua,mate,

va’eribut



I . . . and more marginally with conflicting imperatives:

(29) ?Tumakborun.Imper

kaDir.2sg

daw,Rep

perobut

(dapat)should

huwagNeg.Imp

ko-ngIndir.2sg-Lnk

tumakbo!run‘They say to run, but don’t run!’ [Tagalog]


Reportative exceptionality

I In §4, I analyze such cases as instances of ‘reportativeexceptionality’ in declaratives. (AnderBois, 2014)

(30) a. Pay-kuna-s(s)he-Pl-Rep

noqa-ma-qaI-Illa-Top

qulqi-tamoney-Acc

muntu-ntin-pilot-Incl-Loc

saqiy-wa-nleave-1O-3p = ‘They leave me a lot of money, reportedly . . . ’

b. mana-manot-Impr

rikiright

riku-sqa-ykisee-PP-2

ninot

unone

sol-taSol-Acc

centavo-ta-piscent-Acc-Add

saqi-sha-wa-n-chuleave-Prog-1O-3-Negq = ‘. . . (but) that’s not true, as you have seen, they don’t leaveme one sol, not one cent.’ [Cuzco Quechua], (Faller, 2002, p. 191)


Reportative exceptionality

I In §4, I analyze such cases as instances of ‘reportativeexceptionality’ in declaratives. (AnderBois, 2014)

(31) a. Ma’Neg

bin

Rept-inPfv-A1

maans-(aj)-epass-Status-Def

eeksaamen=o’exam=Distal

‘I didn’t pass the exam reportedly, . . . ’

b. ba’ale’but

t-uPrep-A3

jaaj-il=e’true-Rel=Top

t-inPfv-A1

maans-aj.pass-Status

‘. . . but actually I passed.’ [Yucatec Maya]

§4 An illocutionary account of ImpRep

ImpRep in the discourse scoreboard

I Empirically, I have argued that ImpRep produce the sameillocutionary potential as other imperatives, di↵ering in theaddition of the report and pragmatic e↵ects it brings.

I This section:

I Develop a scoreboard semantics in the style of Farkas & Bruce(2010) for declaratives and imperatives.

I Show that the illocutionary account of reportatives inAnderBois (2014) can be extended to capture ImpRep in auniform way.

§4 Declaratives in the scoreboard

Declaratives contribute ‘dual’ updates

I Uttering a declarative sentence produces two e↵ects/twoupdates to the discourse scoreboard:

Public commitment: Publicly commit a to having adequate evidencesupporting p (Add p to DCa).

Essential e↵ect: Propose to add p to a and b’s Common Ground subjectto b’s approval (Add p to Table{a,b})

(32) Inputa Shared {a,b} b

Table{a,b}DCa CG{a,b} DCb

I Various independent arguments for both of these two updatesin recent literature (e.g. Farkas & Bruce (2010), McCready (2015),

AnderBois (2016), AnderBois (accepted))






(32) Propose and Commita Shared {a,b} b

Table{a,b} p

DCa p CG{a,b} DCb








(32) Accepta Shared {a,b} b

Table{a,b}DCa p CG{a,b} p DCb








(32) Outputa Shared {a,b} b

Table{a,b}DCa p CG{a,b} p DCb



§4 Reportatives as update modifiers

Declarative w/ reportative updates

I AnderBois (2014): illocutionary reportatives in declarativeslike (33) leave the proposal intact, but alter the publiccommitment (see Faller (2002), Murray (2010), Murray (2014) for related

accounts)

(33) Scenario: I am talking on the phone with a friend to ask about theweather in our town so we can decide whether to bring umbrellas andI tell you:TaantikImm.Past

bin

RepuA3

chuunulstart

uA3

toosolsprinkle

ja’=e’water=Top

‘It just started to sprinkle (he says).’

(34) Declarative w/ Reportative bin:a Shared {a,b} b





accounts)


bin

RepuA3

chuunulstart

uA3

toosolsprinkle

ja’=e’water=Top



Table{a,b} p

DCa Rep(p) CG{a,b} DCb




accounts)


bin

RepuA3

chuunulstart

uA3

toosolsprinkle

ja’=e’water=Top



Table{a,b}DCa Rep(p) CG{a,b} p DCb

§4 Imperatives are about preferences

Imperatives and e↵ective preferences

I Much recent literature: imperative meanings involvepreferences.

I However, ordinary preferences like those expressed by verbslike want need not be consistent:

(35) Context: I don’t want to teach at all next semester.

I want to teach Tuesdays and Thursdays next semester. (Heim (1992))

I For imperatives, then, what we need are what Condoravdi &Lauer (2012) call effective preferences.

I i.e. preferences with conflicts resolved in order to guide actions.

§4 Imperatives as proposals

Imperatives in the scoreboard

I To account for imperatives in a parallel fashion: addCPref{a,b} to scoreboard (mutual/common e↵ectivepreferences).

I Parallel essential e↵ect: put a preference for p > ¬p on theTable, proposing to add it to CPref{a,b} if accepted.

(36) Imperative update:a Shared {a,b} b


CPref{a,b}







CPref{a,b}







CPref{a,b}






Table{a,b} p > ¬pDCa CG{a,b} DCb

CPref{a,b}







CPref{a,b} p > ¬p







CPref{a,b} p > ¬p

I (See Starr (2013), von Fintel & Iatridou (t.a.) for similarideas in di↵erent frameworks)


ImpRep updates

I Extending the above to ImpRep, I analyze them as leaving theproposal untouched, but adding the information that thereporter has the given e↵ective preference:


bin

Repayour

wo’ochmeal


‘Drink your pozole (she orders)!’

(38) Imperative w/ Reportative bin:a Shared {a,b} b


CPref{a,b}


ImpRep updates



bin

Repayour

wo’ochmeal




Table{a,b} p > ¬pDCa Rep(p > ¬p) CG{a,b} DCb

CPref{a,b}


ImpRep updates



bin

Repayour

wo’ochmeal




Table{a,b}DCa Rep(p > ¬p) CG{a,b} DCb

CPref{a,b} p > ¬p


ImpRep updates



bin

Repayour

wo’ochmeal




Table{a,b}DCa Rep(p > ¬p) CG{a,b} DCb

CPref{a,b} p > ¬p


Reasoning about speech acts

I As in other recent accounts of imperatives (e.g. Condoravdi &Lauer (2012), Kaufmann (2012)), variability in illocutionaryforce is due to pragmatic reasoning:

I What (likely private) preferences did the various agentspreviously have?

I What is the relative social or rational authority of the agents?

I The speaker’s having committed to a third-party’s e↵ectivepreference in ImpRep influences these pragmatic processes:

I Who is the likely reporter? What is their relativesocial/rational authority?

I Finally, with respect to conflicting imperatives, I followAnderBois (2014) in taking this to be due a pragmaticperspective-shift.

§5 Summary

Empirical claims

I First detailed exploration of ImpRep in discourse, drawing ondata from Yucatec Maya and Tagalog.

I Contrary to several recent authors, I have argued that suchsentences have similar illocutionary potential to otherimperatives (i.e. are ‘imperatives by proxy’ rather than‘neutral reports’).

I Two apparent di↵erences have been argued to be attributableto more general pragmatic processes:

I ‘Softening’ of directive force due to reasoning aboutimperative polyfunctionality, which imperatives show whetheror not they include evidentials.

I Explicit deniability due to pragmatically-driven perspectiveshift, which reportatives facilitate across sentence types.

§5 Illocutionary account needed

Illocutionary vs. modal evidentials

I Well-known dichotomy between illocutionary and modalevidentials (e.g. Faller (2006), Matthewson et al. (2007), Murray (2010))

I i.e. evidentiality ‘above’ and ‘below’ the illocutionary update

I Central to these debates have been the interactions betweeninterrogative sentences and di↵erent evidentials acrosslanguages.

I The illocutionary stability of imperatives with reportativesprovides strong support to illocutionary analyses of these usesof evidentials.

§5 Against the dichotomy

No correlation with other aspects of reportatives

I However, bin and daw di↵er in many other ways, calling thisdichotomy into question: embeddability (e.g. Schwager (2010),

Kierstead (2015)) and felicity in ‘interrogative flip’ scenarios:

(39) Scenario: I know your friend went to a concert and you didn’t, butyou probably talked to your friend about it. I’m asking you about theconcert expecting your answer will be based on the friend’s report.Magalinggood

baPolQ

daw

Repyungthat.Lnk

kumanta?singer

‘According to what you’ve heard, was the singer good?’ [Tagalog]





(40) Scenario: know your friend went to a concert and you didn’t, but youprobably talked to your friend about it. I’m asking you about theconcert expecting your answer will be based on the friend’s report.#Ma’alob-chaj=aagood-Proc=PolQ

leDef

cha’anevent

bin=o’?Rep=Distal

#Intended: ‘According to what you’ve heard, did the event go well?’[Yucatec Maya]





(40) Scenario: know your friend went to a concert and you didn’t, but youprobably talked to your friend about it. I’m asking you about theconcert expecting your answer will be based on the friend’s report.#Ma’alob-chaj=aagood-Proc=PolQ

leDef

cha’anevent

bin=o’?Rep=Distal

#Intended: ‘According to what you’ve heard, did the event go well?’[Yucatec Maya]

I Conclusion: such data call into question the viability of thedichotomy, or at least suggest that individual evidentials mayhave both types of uses (see Korotkova (2016) for a similar conclusion)

§5 Imperatives di↵er in embeddability too

No correlation with other aspects of imperatives

I Schwager (2010), Thomas (2014) suggest that the felicity ofImpRep is related to embeddability of imperatives generally.

I Imperatives in Tagalog and Mbya claimed to be embeddable(Schwager (2010), Thomas (2014))

I However, Yucatec Maya ones are not, most clearly seen forintransitives, which have an imperative su�x -en:

(41) *T-uPfv-A3

{tuklaj/ya’alaj}think/say

[xeen]!go.Imp

Intended: ‘He thought/said to go.’

Jach dyoos bo’otike’ex! Maraming salamat po!

Thank you!

Thanks Rosa Isela Canche Cen, Miguel Oscar Chan Dzul, MargaritaHau Hau, Norma Patricia Kuyoc Kuyoc, Irma Yolanda PomolCahum, and Luis Petul for Yucatec Maya judgments and thoughts!Thanks to Henrison Hsieh, Jenny Tan, and Amber Teng for theirTagalog judgments and thoughts. Thanks also to Polly Jacobson,Kyle Rawlins, 6 anonymous SALT 27 reviewers, and audiences atSAIL 2017 and Brown’s LingLangLunch for helpful feedback.

References: I

Aikhenvald, Alexandra (2004) Evidentiality. Oxford University Press.

Aikhenvald, Alexandra (2010) Imperatives and commands. Oxford UniversityPress.

AnderBois, Scott (2014) On the exceptional status of reportative evidentials. InProceedings of Semantics and Linguistic Theory (SALT) 24, 234–254.

AnderBois, Scott (2016) Semantics and pragmatics of (not-)at -issueness inYucatec Maya attitude reports. Semantics & Pragmatics 9(19).

AnderBois, Scott (accepted) Illocutionary revelations: Yucatec Maya bakaanand the typology of miratives.

Condoravdi, Cleo & Sven Lauer (2012) Imperatives: meaning and illocutionaryforce. In Empirical Issues in Syntax and Semantics 9.

Faller, Martina (2002) Semantics and pragmatics of evidentials in CuzcoQuechua. Ph.D. thesis, Stanford.

Faller, Martina (2006)Evidentiality below and above speech acts, ms. Univ. of Manchester, online athttp://semanticsarchive.net/Archive/GZiZjBhO/Faller-evidentiality.above.below.pdf.

References: II

Faller, Martina (2007) The Cusco Quechua Reportative Evidential andRhetorical Relations. Linguistiche Berichte 14: 223–252.

Farkas, Donka & Kim Bruce (2010) On reacting to assertions and polarquestions. Journal of Semantics 27(1): 81–118.

von Fintel, Kai & Sabine Iatridou (t.a.) A modest proposal for the meaning ofimperatives, ms, submitted to a volume on Modality across syntacticcategories (Oxford University Press), ed. by Ana Arregui, Marisa Rivero, andAndres Pablo Salanova.

Hamblin, C.L. (1987) Imperatives. Basil Blackwell.

Hanks, William (1990) Referential Practice: Language and Lived Space amongthe Maya. University of Chicago Press.

Heim, Irene (1992) Presupposition projection and the semantics of attitudeverbs. Journal of Semantics 9: 183–221.

Kaufmann, Magdalena (2012) Interpreting imperatives. Springer.

Kierstead, Gregory (2015) Projectivity and the Tagalog reportative evidential.Master’s thesis, The Ohio State University.

References: III

Korotkova, Natalia (2016) Heterogeneity and uniformity in the evidentialdomain. Ph.D. thesis, UCLA.

Korotkova, Natasha (2015) Evidentials and (relayed) speech acts: Hearsay asquotation. In Proceedings of SALT 25, 676–694.

Laughren, Mary (1982) A preliminary description of propositional particles inWarlpiri. In Work Papers of the Summer Institute of Linguistics, AustralianAborigines Branch A 6, SIL, 129–163.

Lim, Dong Sik (2010) Evidentials and interrogatives: a case study fromKorean. Ph.D. thesis, USC.

Lucy, John (1993) Metapragmatic Presentationals: Reporting Speech withQuotatives in Yucatec Maya. In Reflexive Language: reported speech andmetapragmatics, John Lucy, ed., Cambridge University Press, 91–125.

Matthewson, Lisa, Henry Davis, & Hotze Rullman (2007) Evidentials asepistemic modals: evidence from St’at’imcets. Linguistic Variation Yearbook7: 201–254.

McCready, Eric (2015) Reliability in pragmatics. Oxford University Press.

References: IVMurray, Sarah (2010) Evidentiality and the structure of speech acts. Ph.D.

thesis, Rutgers.

Murray, Sarah (2011) A Hamblin semantics for evidentials. In Proceedings fromSemantics and Linguistic Theory (SALT) XIX (2009), CLC Publications,324–341.

Murray, Sarah (2014) Varieties of update. Semantics & Pragmatics 7(2): 1–53.

Schachter, Paul & Fe T. Otanes (1972) Tagalog Reference Grammar.University of California Press.

Schwager, Magdalena (2010) On what has been said in Tagalog. In Evidencefrom evidentials, University of British Columbia Working Papers inLinguistics, 221–246.

Starr, William (2013) A preference semantics for imperatives, ms. Cornell.

Thomas, Guillaume (2014) Embedded imperatives in Mbya. In Proceedings ofNELS 43, Graduate Linguistic Student Association of the University ofMassachusetts, Amherst, 181–194.

Wilkins, David P. (1989) A grammar of Mparntwe Arrernte. Ph.D. thesis,Australian National University.

EXTRA SLIDES

Extra slides

§A Reportatives vs. quotatives

Quotatives are particles with reporters encoded


I Whereas quotatives allow for the reporter to be expressedexplicitly as an argument or via agreement, as in (42), . . .

I . . . reportatives do not allow for an overt reporter argument:

(42) Scenario: I was talking to my friend Luis earlier about the Xtaabayand now tell you:Luis=e’Luis=Top

chowaklong

uA3


uA3

poolhead

leDef

ixtaabay=o’Xtaabay=Distal

–kij

Quot.B3sg(teen).Dat.1sg

‘As for Luis, “the Xtaabay’s hair is long.” he tells me. [Yucatec Maya]

(44) Scenario: I was talking to my friend Luis earlier about the Xtaabayand now tell you:*Luis=e’Luis=Top

chowaklong

bin

RepuA3


uA3

poolhead

leDef


§A Reportatives vs. quotatives

Quotatives are particles with reporters encoded


I Whereas quotatives allow for the reporter to be expressedexplicitly as an argument or via agreement, as in (42), . . .

I . . . reportatives do not allow for an overt reporter argument:

(43) Scenario: I was talking to my friend Luis earlier about the Xtaabayand now tell you:*Luis=e’Luis=Top

chowaklong

bin

RepuA3


uA3

poolhead

leDef


§A Comparison with quotative k- in YM

Quotative k- in YM

I Within the neutral report approach, Korotkova (2015),Korotkova (2016) explicitly regard ImpRep as quotations.

I YM has grammaticized ‘parenthetical’ quotative marker k(ij)-,allowing for straightforward comparison (Lucy (1993), Hanks (1990))

I Unlike both verbs of saying and bin, k(ij)- occurs onlyfollowing the quoted material or prosodically separated inmedial position.

I Like verbs of saying, but unlike bin, k(ij)- inflects to agree withthe reporter and allows the recipient of the report to beexpressed with a dative.

(44) Tu’uxwhere

k-aImp-A2

bingo

- k-enQuot-B1sg

ti’.Dat.3sg

‘ “Where are you going?” I said to him.’ [Yucatec Maya] (Lucy, 1993)

§A Pronouns and shifty things in quotatives

Pronouns and shifty things in quotatives

I Unlike reportatives, local pronouns and other deicticexpressions shift with k(ij):

(45) Wayhere

k-inImp-A1

kutal-e’sit-Top

- kijQuot.3sg

‘ “I’ll sit here” he says.’ (He’s like ‘I’ll sit here).’ [Yucatec Maya]

(Hanks, 1990, p. 207)

I The implicit imperative subject similarly shifts and the resultis a neutral report or quote:

(46) Scenario: A friend is asking what my father (who he hasn’t talked to)told me to do:Pee(k)n-enmove.Antip-Imper

EspitaEspita

- kijQuot

‘ “Go to Espita!” he said.’ [Yucatec Maya]

§A Illocutionary force of quotatives

Imperatives with quotatives really are ‘neutral reports’

I This is so regardless of how the arguments of k(ij) are spelledout:

(47) Xeengo.Imper

- kijQuot.3sg

tech.Dat.2sg

‘ “Go”, he said/says to you’ [Yucatec Maya]

(48) Ko’ox=i’go.Hort=Deic

- kiQuot.3sg

teenDat.1sg

‘ “Let’s go!” he said to me’ [Yucatec Maya]

I While both bin Rep with k(ij)- Quot ‘pass along’ the speechof another agent, only the latter produces neutral reports.

I Lucy (1993): k(ij) is “functionally restricted tometacommunicative uses”

Date post:	08-Mar-2018
Category:	Documents
Upload:	hakien
View:	224 times
Download:	2 times

An illocutionary account of reportative evidentials in...

Documents