Why are 'you' and 'I' so special? Stephen Wechsler [email protected] University of Texas at Austin

UNC, Chapel Hill January 13, 2012

1

personal pronouns distinct pronoun forms depending on the speech act role of the referent: speaker:

1st person

I, me, my, we, us, our…

addressee:

2nd person you, your, y’all, y’all’s…

other:

3rd person

he, him, his, she, her, it, its, they, them, their…

where ‘other’ includes overhearers and non-users

2

How are personal pronouns specified? Two alternative views of personal pronouns: • Descriptive content theories: 1st / 2nd person pronouns are specified for reference to the speaker / addressee (Kaplan 1977; Schlenker 2003; Maier 2010; Anand and Nevins 2004) • Self-identification theory: 1st / 2nd person pronouns are specified for self-identification by the speaker / addressee (Wechsler 2010) ⇐ projected winner

3

How are personal pronouns specified? Two alternative views of personal pronouns: • Descriptive content theories: 1st / 2nd person pronouns are specified for reference to the speaker / addressee (Kaplan 1977; Schlenker 2003; Maier 2010; Anand and Nevins 2004) • Self-identification theory: 1st / 2nd person pronouns are specified for self-identification by the speaker / addressee (Wechsler 2010) ⇐ projected winner

4

Contents 1. Semantics of personal pronouns a. Kaplan’s (1977) theory b. Self-identification theory 2. Context-shifting 3. The Associative Plural Generalization 4. Conjunct/disjunct marking 5. Autism 6. Conclusion

5

Indexicals With indexicals, ‘the referent is dependent on the context of use and the word provides a rule which determines the referent in terms of certain aspects of the context’ (Kaplan 1977 p. 490) examples: I, me, my, you, your, that, this, here, now, tomorrow, yesterday, actual, present, local, …

6

Role of the descriptive content of indexicals [the president of PepsiCo] = ‘the person serving as corporate

head of the company called PepsiCo’ This descriptive content can be relevant to determining the relevant individual in the circumstance of evaluation [I] = ‘the current speaker’

This descriptive content is relevant only to determining a referent in a context of use and not to determining a relevant individual in a circumstance of evaluation (Kaplan 1977)

7

De dicto vs. de re belief 1. John believes the president of PepsiCo is rich. de dicto: John believes the proposition that whoever is the president of PepsiCo is rich (on the basis of, say, his general belief that presidents of large corporations are always rich) n.b: Does not require that John or the reporter have any idea who the president of PepsiCo is paraphrase: ‘John believes the president of PepsiCo, whoever that may be, is rich.’ (Example from Maier 2010)

8

De dicto vs. de re belief reporting 1. John believes the president of PepsiCo is rich. de re: John believes of a certain individual, who is the actual president of PepsiCo, that she is rich. On this de re construal, the report is felicitous if John just talked to Indra Nooyi at a party about her yacht, and believes ‘‘wow, this woman is rich’’. paraphrase: ‘John believes of the president of PepsiCo that she is rich.

9

indexicals like ‘I’: only de re 3. Indra Nooyi says: “John believes that I am poor.” In the world of John’s beliefs: Indra Nooyi is poor. (‘John believes of Indra Nooyi that she is poor.’) NOT: A poor person is speaking, instead of Indra Nooyi. (‘John believes that the person who is currently speaking, whoever that may be, is poor.’) [I] = ‘the current speaker’. But this descriptive content is

relevant only to determining a referent in a context of use and not to determining a relevant individual in a circumstance of evaluation 10

indexicals like ‘I’ and ‘you’: only de re 4. Necessarily, I am speaking now. ‘It is necessarily the case that I am speaking now.’ (false!) [NOT: ‘It is necessarily the case that the current speaker is speaking now.’ (true!)] [I] = ‘the current speaker’. But this descriptive content is

relevant only to determining a referent in a context of use and not to determining a relevant individual in a circumstance of evaluation

11

An example from Kaplan 1. It is possible that in Pakistan, in five years, only those who are actually here now are envied. In the possible world described by this utterance: here = place of utterance

(here ≠ in Pakistan)

now = time of utterance

(now ≠ five years hence)

12

Kaplan’s solution Step 1. Evaluate all indexicals w.r.t. the utterance context. Barack Obama says: “I am in Austin.” rule: substitute the speaker for ‘I’ => Barack Obama is in Austin. in-Austin(B.O.)

Step 2. Determine the extension (for a statement, its truth value) w.r.t. circumstances of evaluation. in-Austin(B.O.) in Circumstance#1: True in-Austin(B.O.) in Circumstance#2: False in-Austin(B.O.) in Circumstance#3: False

…etc. 13

Kaplanian utterance context A Kaplanian context: a tuple of various semantically relevant parameters c = 〈 sp, h, loc, t, … 〉 indexicals have denotations which depend directly on such contexts in order for their meaning to be instantiated: [| I |] = sp(c) [| you |] = h(c) [| here |] = loc(c) [| now |] = t(c)

14

Kaplan on the descriptive content of indexicals ‘I am not claiming… that indexicals lack anything that might be called ‘descriptive meaning.’ Indexicals, in general, have a rather easily statable descriptive meaning. But it is clear that this meaning is relevant only to determining a referent in a context of use and not to determining a relevant individual in a circumstance of evaluation.’ (Kaplan 1977, 498)

15

Self-identification theory (Wechsler 2010) • 1st / 2nd person pronouns lack the descriptive content ‘speaker’ and ‘addressee’ entirely— even in the context of use. Instead: • the speaker’s rule of use for 1st person pronouns is: self-identify with the pronoun referent. • the addressee’s rule of use for 2nd person pronouns is: self-identify with the pronoun referent.

16

identity as an acquaintance relation Lewis (1979): To have a de re belief (a belief about a real thing), one must bear some acquaintance relation with it. Self-identification (belief de se): belief de re, in which the acquaintance relation is identity.

Lewis, David 1979. ‘Attitudes de dicto and de se.’

17

de re belief requires acquaintance with the res 1. John believes the president of PepsiCo is rich. de re: John believes of a certain individual, who is the actual president of PepsiCo, that she is rich. On this de re construal, the report is felicitous if John just talked to Indra Nooyi at a party about her yacht, and believes ‘‘wow, this woman is rich’’. paraphrase: ‘John believes of the president of PepsiCo that she is rich.

18

relational semantics of de re belief de re belief: the subject must be vividly acquainted with a res (Lewis 1979) x believes de re of y that y has the property P iff there is a relation R where: • R is a vivid relation of acquaintance between x and y in the actual world • x believes the proposition that the object he is Racquainted with has the property P. In our example the ‘relation of acquaintance’ between John and the res is: meeting her at a party.

19

de re and de se 1. Perryi believes hei is making a mess. a. de re (but not de se): Perry sees himself in the mirror; doesn’t realize it is himself that he sees ⇒ vivid acquaintance relation R: see-in-mirror b. de se: Perry is in a position to say ‘I am making a mess.’ ⇒ vivid acquaintance relation R: identity

20

de re and de se (a little more formally) object of belief: a triple 〈 a, R , P 〉 : an agent a, an individual x, an acquaintance relation R(a,x), and a property P. 1. Johni believes hei is making a mess. de re: Bel( j, 〈 j, see.in.mirror(j,x) , λy.mess.maker(y)〉 ) de se: Bel( j, 〈 j, j = x , λy.mess.maker(y)〉 )

21

From objects of beliefs to unembedded sentences Not only beliefs but also assertions and interpretations involve individuals under acquaintance relations. (McCready and Wechsler 2012) Assertions are made based on the knowledge we have: to sincerely predicate some property of an individual, one must be acquainted with her in some manner. Interpretation by an addressee, similarly, requires acquaintance with the res. Model with centered worlds: world-agent pairs (Lewis 1979, Stalnaker 2008, Maier 2009). 22

The self-identification theory of 1st/2nd person Speaker’s rule for the use of I: Use I to indicate that Ψ is instantiated to identity. Addressee’s rule for interpreting you: Interpret you as indicating that R is instantiated to identity.

(McCready and Wechsler 2012)

23

I = self

I like you

Mary

Paula

Speaker’s rule for the use of I: Use I to indicate that R is instantiated to identity.

24

I = self you = self I like you

Mary

Paula

Addressee’s rule for you: Interpret you as indicating that R is instantiated to identity.

25

I = self you = self

I like you.

Mary

Paula

How speaker interprets ‘you’: she builds a model of the addressee’s mental state ( ≈ Gricean pragmatics) Theory of Mind (ToM): the ability to impute mental states to others (Premack and Woodruff, 1978). 26

I = self you = self

I like you.

Mary

Paula

How addressee interprets ‘I’: she builds a model of the speaker’s mental state ( ≈ Gricean pragmatics) Theory of Mind (ToM): the ability to impute mental states to others (Premack and Woodruff, 1978). 27

Theory comparison Descriptive content theories: The notions ‘speaker of c’ and ‘addressee of c’ are used for ‘determining a referent in a context of use’. (Kaplan 1977, 498). Includes Neo-Kaplanian theories: Schlenker, Anand, Maier, Hunter. Self-identification theory: The notions ‘speaker of c’ and ‘addressee of c’ are not used for determining a referent. They are relevant only to determining who must follow the rule. The notion relevant to determining a referent is ‘selfidentification’.

28

An analogy Gender-Exclusive Differentiation: male & female speakers have different speech forms. Koasati (Muskogean; Haas 1944:143): • for a female speaker: • for a male speaker:

tačílw tačílws

‘you are singing’ ‘you are singing’

‘female’ is not part of the descriptive content of tačílw ‘male’ is not part of the descriptive content of tačílws The descriptive content of both forms is ‘you are singing’. The notions ‘male’ / ‘female’ are relevant only to determining who must follow the rule. 29

Similarly for 1st/2nd person: The speaker or addressee self-identifies with the pronoun. • The semantic content of ‘I’ and ‘you’ is the same: both are for self-identification. • They differ in who is permitted to self-identify with them: the speaker or the addressee, respectively.

30

Evidence for the self-identification theory: • Context-shifting • The Associative Plural Generalization • Conjunct/disjunct marking • Autism

31

Context-shifting Most indexicals allow shifting from the utterance context to a reported context (exs. from Hunter 2010, 2011): 1. Now he was back in Africa, without his wife and son. 2. The writer was Lucy Mercer Rutherfurd, who decades before had been FDR’s mistress and who now was making arrangements for what would be their last fateful meeting at the president’s rural retreat. 3. He built a house by the landing-place on the island for his visitors to stay in, and here, too, his monks would come on festivals to have a talk with him.

32

Neo-Kaplanian approaches to shifting Allow context to vary. 1. Now he was back in Africa, without his wife and son. utterance context: reported context:

c c’

[| here |] = loc(c) [| now |] = t(c)

(utterance location) (utterance time)

[| here |] = loc(c’) [| now |] = t(c’)

(Africa) (time of his return to Africa) 33

But I and you do not shift • No unembedded shifting, as in 1-3, for I and you. • Nor does shifting occur in embedded contexts: 5. a. John felt that now he could call Mary. b. John felt that now I could call Mary. (I ≠ John) 6. a. John built a house by the landing-place and decided that here, too, his monks could visit him. b. John built a house by the landing-place and decided that here, too, his monks could visit me. (me ≠ John)

34

Context-shifting ≠ deferred reference Nunberg (1993:20): Sometimes an indexical pronoun behaves like a description instantiated by the index: A condemned prisoner says: ‘I am traditionally allowed to order whatever I like for my last meal.’ = ‘A condemned prisoner is traditionally allowed to order whatever he likes for his last meal’. • index of I: the speaker N.b. This cannot be shifted! • description instantiated by the index: ‘condemned prisoners’

35

Let us assume a neo-Kaplanian account of shift for time, loc indexicals: replace c with c’: [| here |] = loc(c’) [| now |] = t(c’) Q: Why don’t ‘you’ and ‘I’ shift? On the neo-Kaplanian theory of you/I: it is lexically stipulated that they don’t shift, in contrast to other indexicals. On the self-identification theory of you/I: There are no rules like [| I |] = sp(c) or [| you |] = hr(c). Hence no possibility of shifting.

36

shifted indexical languages Amharic: 1. Situation: John says: ‘I am a hero’ ǰon ǰəәgna nəә-ññ yɨl-all. John hero be.PF-1SO 3M.say-AUX.3M ‘John says that he is a hero.’ 2. [min amt'-a ind-al-əә-ññ] al-səәmma-hu-mm what bring.IMPER-2M COMP-say.PF-3M-1SO NEG-hear.PF-1S-NEG ‘I didn’t hear what he told me to bring.’ (lit. ‘I didn’t hear that he said to me, you bring what.’) (Leslau 1995 p. 779) Not direct quotations: if 2 were a quotation, the original discourse should have been of the form: ‘Bring what!’, which doesn’t make sense. 37

Analysis of shifted indexical languages: the 1st & 2nd person forms, like the shiftable time and place indexicals, have a Neo-Kaplanian analysis. The self-identification theory predicts an asymmetry: speech participant indexicals are the last to allow shifting.

38

The Associative Plural Generalization • Speech act roles: speaker, addressee • What distinctions are drawn by personal pronoun systems? • What does the answer tell us about the semantics of person?

39

1st and 2nd person

I= you.SG =

‘speaker’ ‘addressee’ (addr.)

we = ‘speaker + speaker’ ‘speaker + addr.’ ‘speaker + other’ ‘speaker + addr. + other’ you.PL = ‘addr. + addr.’ ‘addr. + other’

We are the champions! Shall we go? We want you to join us. Can’t we all get along?

You should behave yourselves. Who did you guys hire? (1 addr.)

40

1st Person Plural: Inclusive vs. Exclusive • inclusive (of addressee): ‘you & I (& possibly others)’ Indonesian:

kita

• exclusive (of addressee): ‘I & others’ (not you) Indonesian:

kami

41

person systems English ‘speaker’ ‘addressee’

I you

Indonesian saya kamu

‘speaker + speaker’ ‘speaker + other’ ‘speaker + addr.’ ‘speaker + addr. + other’

we we we we

kami kami kita kita

‘addr. + addr.’ ‘addr. + other’

you you

kalian kalian

42

person systems English ‘speaker’ ‘addressee’

I you

Indon. saya kamu

‘speaker + speaker’ ‘speaker + other’ ‘speaker + addr.’ ‘speaker + addr. + other’

we we we we

kami kami kita kita

‘addr. + addr.’ ‘addr. + other’

you you

kalian kalian

43

Unattested

kamu kama

person systems English ‘speaker’ ‘addressee’

I you

Indon. saya kamu

‘speaker + speaker’ ‘speaker + other’ ‘speaker + addr.’ ‘speaker + addr. + other’

we we we we

kami kami kita kita

‘addr. + addr.’ ‘addr. + other’

you you

kalian kalian

44

Unattested

kitu kiti

person systems English ‘speaker’ ‘addressee’

I you

Indon. saya kamu

‘speaker + speaker’ ‘speaker + other’ ‘speaker + addr.’ ‘speaker + addr. + other’

we we we we

kami kami kita kita

‘addr. + addr.’ ‘addr. + other’

you you

kalian kalian

45

Unattested

kaliu kalia

Seven ‘meta-persons’; only four attested pronoun types Possible 1+2

Attested

speaker(s) and addressee(s) only

1+2+3 speaker(s), addressee(s), & other(s) 1 speaker(s) only (‘true 1PL’) 1+3

speaker(s) & other(s) only

2

addressee(s) only (‘true 2PL’)

2+3

addressee(s) & other(s) only

3

other(s) only

‘inclusive’ ‘exclusive’ ‘second person’ ‘third person’

46

The Associative Plural Generalization U1. No language distinguishes [1+1] from [1+3]. U2. No language distinguishes [2+2] from [2+3]. U3. No language distinguishes [1+2] from [1+2+3]. (Bobaljik 2008) ‘However great the semantic plausibility, the category 2+2 is not found grammaticalized in the languages of the world.’ (Cysouw 2003:75). (Bobaljik, 2008; Cysouw, 2003; Greenberg, 1988; McGregor, 1989; Moravcsik, 1978; Noyer, 1992; Silverstein 1976)

47

The APG: a typological universal? Bobaljik’s 2008 metastudy of large-scale typological studies: ‘the empirical basis is extremely well-documented’ (207) ‘sample sizes on the order of 500 languages’ from studies over the past half century (Forchheimer 1953; Sokolovskaja 1980; Cysouw 2003). ‘absolute universals rather than strong trends’ (209) Cysouw 2003: 309 languages. Siewierska (2004, 82-3) concurs.

48

The APG is semantically strange Full NPs: regular plural is the norm; some associatives.

regular plural NP: ‘Every member of reference set is…’ the dogs: ‘Every member of the reference set is a dog.’ associative NPs: ‘Some member of reference set is…’ Péter-ék: ‘Some member of the reference set is named Péter.’ (Hungarian) The APG: 1st /2nd PL are associative in all languages.

49

Explanation for the Associative Plural Generalization

On the Self-Identification theory: Speaker and addressee are not part of descriptive content of 1st and 2nd person pronouns, so such pronouns cannot be restricted to just speakers/addressees. On the Descriptive Content theory: APG is a mystery. If you.SG refers to ‘addressee’ (in the context), then why are there no plural pronouns referring to ‘addressees’ (in the context)?

50

2nd sg, multiple addressees Teacher to her class: Write your name on the paper. Each addressee x knows to write x’s name: Tom writes Tom’s name; Mary writes Mary’s name. Lesson: 2nd Person induces self-identification by each addressee, not reference to addressees.

51

2nd pl, multiple addressees Addressing a marriage workshop for husbands (no wives present), asking about their respective marriages: How often do you kiss each other? Each addressee x interprets you as referring to a set that includes x (plausibly, x and x’s spouse) Lesson: you does NOT refer to the set of addressees, nor to a superset of the addressees.

52

Conjunct/disjunct marking In some languages the verb form, Conjunct (CJ) or Disjunct (DJ), depends on person and illocutionary type, as follows: Declarative Interrogative 1st CJ DJ 2nd DJ CJ 3rd DJ DJ

53

Newari (Sino-Tibetan; Hargreaves 2005) jī:     a:pwa   twan-­‐ā.   1.ERG   much   drink-­‐PST.CJ   ‘I  drank  too  much.’  

jī:   a:pwa  twan-­‐a   lā?   1.ERG  much   drink-­‐PST.DJ   Q   ‘Did  I  drink  too  much?’     chā   a:pwa   twan-­‐ā   lā?   2.ERG  much   drink-­‐PST.CJ  Q   ‘Did  you  drink  too  much?’     wā:   a:pwa  twan-­‐a   lā?   3.ERG  much   drink-­‐PST.DJ   Q   ‘Did  s/he  drink  too  much?

chā     a:pwa   twan-­‐a.   2.ERG   much   drink-­‐PST.DJ   ‘You  drank  too  much.’     wā:   a:pwa   twan-­‐a.   3.ERG   much   drink-­‐PST.DJ   ‘S/he  drank  too  much.’    

54

de se reports (self-reports) also use CJ:

   

Syām-­‐ā   a:pwa   twan-­‐ā   hā.   Syam-­‐ERG   much   drink-­‐PST.CJ   EVD   ‘Syami  said  that  hei  drank  too  much.’  

55

Some languages with CJ/DJ systems Sino-Tibetan: Newari (Hale 1980; Hargreaves 2005) Nakh-Daghestanian: Akhvakh (Creissels 2008) Mehwb Dargwa (Bickel 2008) Trans New Guinea: Oksapmin (Loughnane 2009) Duna and Kaluli (San Roque 2011) Guambiano (Norcliffe 2011) Cha’palaa (Floyd 2011) Tsafiki (Barbacoan; Dickinson 2000)

56

Hypothesis: CJ verb morphology marks a verb phrase as denoting a self-asserted property: a property someone asserts to hold of herself (i.e., under an acquaintance relation of identity). object of assertion: a triple 〈 a, R , P 〉 , i.e., an agent a’s assertion that an individual x, under acquaintance relation R(a,x), has property P. 1. Syami said that hei drank too much.

Assert( s, 〈 s, s = x , λy.drank.too.much(y)〉 )

57

Uses of Conjunct Marking • A declarative utterance is a declaration of speaker belief. In 1st person subject CJ declaratives the speaker selfasserts the property denoted by the CJ-marked VP. • In 2nd person subject CJ questions the addressee is asked whether she is prepared to self-assert the property denoted by the CJ-marked VP. • Reported speech using CJ reports someone’s selfassertion. Lesson: self-identification is a component of human activity, not a component of semantic content. 58

Personal pronouns and autism • Childhood autism: a developmental disorder characterized by specific social, emotional, cognitive, and linguistic impairments. • While many children with autism are mentally retarded, others have IQ’s within the normal range. (Baron-Cohen et al., 1985; Tager-Flusberg, 2001; TagerFlusberg and Joseph, 2005).

59

Linguistic symptoms of autism Since the earliest descriptions (Kanner 1943), researchers have noted a special difficulty with the use of 1st and 2nd person pronouns, ‘to a degree that seems out of keeping with other aspects of their language development’ (Lee et al., 1994: 156) (Kanner, 1943, Bettelheim, 1967; Fay, 1979; Lee et al., 1994: 156; Tager-Flusberg, 1994: 184).

60

Errors of person Children with autism tend to reverse pronouns. (Tager-Flusberg, 1994: 184): transcripts of the speech of children with autism and Down syndrome, analyzed for person reversal errors (you for I, etc.) and case errors (me for I, etc.). Result: Autism Down syndrome

person reversal 13% (220 / 1,673) 0% (0 / 2,270)

61

case errors 0.12% (2 / 1,673) 1.23% (28 / 2,270)

Why? ‘Autistic children have great difficulty understanding that different people have distinct conceptual perspectives— that people perceive, interpret, remember, and respond to situations in unique ways. This kind of conceptual perspective-taking ability is required in order to understand the different roles of speaker and listener (Loveland 1984), and this is reflected in the way pronouns are used in ongoing discourse.’ (Tager-Flusberg 1994)

62

Theory of Mind Hypothesis of Autism • Attributes autism to a deficit or lack of a ToM. • Evidence from false-belief tests. (Baron-Cohen, Leslie, and Frith 1985; Tager-Flusberg 2001; Tager-Flusberg and Joseph 2005)

63

What does that tell us about the semantics of indexicals? On the (Neo-)Kaplanian theory of indexical pronouns (‘1st person refers to the speaker, in the context of utterance’) there is no need to take someone’s perspective in order to understand a pronoun referring to them. For example, to understand ‘I’, all that one needs to know is (i) who is speaking, and (ii) that a first person pronoun refers to whoever is speaking. Mutatis mutandis for addressees and 2nd person pronouns.

64

On the Self-Identification Theory: • an addressee must comprehend the perspective of the speaker in order to understand a 1st person pronoun. • a speaker must comprehend the perspective of the addressee in order to understand a 2nd person pronoun.

65

I = self you = self

I like you.

Mary

Paula

How speaker interprets ‘you’: she builds a model of the addressee’s mental state (≈ Gricean pragmatics) Theory of Mind (ToM): the ability to impute mental states to others (Premack and Woodruff, 1978).

66

I = self you = self

I like you.

Mary

Paula

How addressee interprets ‘I’: she builds a model of the speaker’s mental state (≈ Gricean pragmatics) Theory of Mind (ToM): the ability to impute mental states to others (Premack and Woodruff, 1978).

67

Conclusion: a larger role for extra-grammatical inference Mary says to Paula: ‘I like you’. It express the proposition that Mary likes Paula. How? What I hope to have shown is that is it not by this translation procedure: I ⤳  ‘speaker’  ⤳ Mary   you ⤳  ‘addressee’  ⤳  Paula     Moreover, there cannot be any such translation procedure.

68

Communication involves extra-grammatical inference based on these rules: • Speaker’s rule: self-identify with the referent of ‘I’. • Addressee’s rule: self-identify with the referent of ‘you’. Other examples of extra-grammatical inference: • calculating conversational implicatures (Grice 1975) • interpreting vague expressives (McCready 2011) • lexical pacts (Brennan and Clark 1996)

69

The proposed application of extra-grammatical inference to semantics of ‘I’ and ‘you’ does not fit into previous applications of it: • unlike Gricean implicature (conversational or conventional), it contributes to literal meaning • unlike conversational implicature, it is conventional • unlike conventional implicature, it is not built into the translation procedure

70

Expressives: express speaker’s attitude 1.

Man, it’s 92 degrees outside.

McCready (2011): expressive indicates speaker’s emotive attitude towards the descriptive content; but the expressive is not translated. Addressee uses nonmonotonic inference to determine the speaker’s attitude. Unlike expressives, I/you: • contribute to descriptive content. • are not vague (expressives often involve underspecified emotive content; McCready 2011). Thus: the mechanism is familiar but the application is new. 71

Assuming • rules of self-identification; and • extra-grammatical inference mechanisms then languages have a means for refering to the speaker and addressee. This obviates the need for pronouns specified for reference to speaker or addressee.

72

73

Thank-you!

74

75

References Anand, Pranav, and Andrew Nevins. 2004. Shifty operators in changing contexts. In Proceedings of SALT XIV, ed. R. Young, 20-37. Ithaca, NY: CLC Publications, Cornell University. Baron-Cohen, Simon, Alan M. Leslie, and Uta Frith. 1985. “Does the autistic child have a ‘theory of mind’?” Cognition 21: 37–46. Bickel, Balthasar. 2008. VerbAgreement and EpistemicMarking: a Typological Journey from the Himalayas to the Caucasus. In Festschrift für Roland Bielmeier, ed. Chomolangma, Demawend, and Kasbek, 1–14. Brennan, S. E, and H. H Clark. 1996. “Conceptual pacts and lexical choice in conversation.” Journal of Experimental Psychology: Learning, Memory, and Cognition 22 (6): 1482. Creissels, Denis. 2008. “Person variations in Akhvakh verb morphology: functional motivation and origin of an uncommon pattern.” STUFLanguage Typology and Universals 61 (4): 309–325.

76

Dickinson, C. 2000. “Mirativity in Tsafiki.” Studies in Language 24 (2) (September): 379-422. doi:10.1075/sl.24.2.06dic. Floyd, Simeon. 2011. Conjunct/disjunct marking in Cha’palaa HANDOUT. Grice, H. P. 1975. Logic and conversation. In Syntax and Semantics: Volume 3, ed. Peter Cole and Jerry L. Morgan, 41-58. New York: Academic Press. Hale, A. 1980. “Person markers: Finite conjunct and disjunct verb forms in Newari.” Papers in South-East Asian Linguistics 7: 95–106. Hargreaves, David. 2005. “Agency and intentional action in Kathmandu Newar.” Himalayan Linguistics 5: 1–48. Kaplan, David. 1977. Demonstratives: An essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals. In Themes from Kaplan, ed. Joseph Almog, John Perry, and Howard Wettstein, 481–566. NY and Oxford: Oxford University Press. Lewis, David. 1979. “Attitudes De Dicto and De Se.” The Philosophical Review 88: 513-543.

77

Loughnane, Robyn. 2009. A grammar of Oksapmin. http://repository.unimelb.edu.au/10187/4788. Maier, Emar. 2010. “Presupposing acquaintance: a unified semantics for de dicto, de re and de se belief reports.” Linguistics and philosophy 32 (5): 429-474. McCready, Eric. 2011. Emotive Equilibria. Draft. McCready, Eric, and Stephen Wechsler. 2012. Why “you” and “I” do not shift. Abstract. University of Texas, Austin. Norcliffe, Elisabeth. 2011. Conjunct/disjunct patterns in Guambiano HANDOUT. San Roque, Lila. 2011. An introduction to conjunct/disjunct alignment in Duna and Kaluli (Trans New Guinea) HANDOUT. Schlenker, Philippe. 2003. “A plea for monsters.” Linguistics and Philosophy 26: 29-120. Tager-Flusberg, Helen. 1994. Dissociations in Form and Function in the Acquisition of Language by Autistic Children. In Constraints on Language Acquisition: Studies of Atypical Children, ed. Helen TagerFlusberg, 175-194. Hillsdale, NJ: Lawrence Erlbaum Associates.

78

———. 2001. A Reexamination of the Theory of Mind Hypothesis of Autism. In The Development of Autism― Perspectives From Theory and Research, ed. Jacob A Burack, Tony Charman, Nurit Yirmiya, and Philip R Zelazo, 173-193. Mahwah, NJ: Routledge. Tager-Flusberg, Helen, and Robert M Joseph. 2005. Theory of Mind, Language, and Executive Functions in Autism: A Longitudinal Perspective. In Young Children’s Cognitive Development: Interrelationships Among Executive Functioning, Working Memory, Verbal Ability, and Theory of Mind, ed. Wolfgang Schneider, Ruth Schumann-Hengsteler, and Beate Sodian, 239-257. Mahwah, NJ: Routledge. Wechsler, Stephen. 2010. “What ‘you’ and ‘I’ mean to each other: Person indexicals, self-ascription, and theory of mind.” Language 86 (2): 332-365.

79