Course 4 REFERENCE AND INFERENCE Words tiemselves do not refer to anything. People refer. Reference is an act in which a speaker/writer uses linguistic forms to enable a listener/reader to identify something. Those linguistic expressions are referring expressions, which can be proper “Touns Byron, Clinton, Romania), noun phrases which are definite (the author, the singer, the artist) or indefinite (a man, a woman, a TICE place), and pronouns (he, she, them). The choice of one type of referring expression over another seems to be based on what the SpeaKer assumes the listener already knows, In shared visual contexas, the pronouns that function as deictic expressions (take this, look at her) may be sufficient for successful reference, but where identification is more difficult, more elaborate noun Phrases may be used (Remember that funny happening when we found a hat?) ener: is thed to the speaker's goals (to identify something} and the speaker's ‘belies (can the listener Know that particular something?) in the use of language—For— Sticcessful reference to occur, the role of inference must be considered. Because there is no direct relationship between entities and words, the listener's task is to infer correctiv ‘which entity the speaker intends to identify by using a particular referring expression, It is not unusual Tor people to Want (6 refer to Some entity or person without knowin exactly what name or word to use. We can even use vague expressions (Ihe Téd thing, that itchy stuff, what’s his name) relying on the listener's ability to infer what referent we have in mind. Speakers even invent names. —— — Reference is not based on an objectively correct versus incorrect naming, but on some locally successful versus unsuccessful choice of expression, Successful reference is necessarily collaborative, with both the speaker and the listener having a role in thinking ‘about what the other has in mind, Names and referents There is a convention that certain referring expressions are used to identify certain entities on a regular basis. Itis our daily experience of the successfull operation of this convention that may cause us to assume that referring expressions can only designate err specific entities. Tis asumption may Tea uso Thnk ta nae or pers foun like ‘Byron’ can only be used to identify one specific person, and an expression containing a common noun, such as “the chicken sandwich,’ can only be used to identify a specific thing. This belief is mistaken. A pragmatic view of reference allows us to see how a person can be identified via the expression ‘the chicken sandwich’ and athing can be identified via the name “Byron.” For example, it is not strange for one student to ask: ‘A: Can I borrow your Byron? B: Yeah, it’s over there on the shelf Given the context just created, the intended referent and the inferred referent would not be a person, but a book Ina restaurant, a waiter brings an order of food for another waiter and asks: A; Where’s the chicken sandwich sitting? B: He's over here by the window. Given the context, the referent being identified is not a thing, but a person. The following examples may allow for a clearer view on how reference actually works:Shakespeare takes the whole top sheit hated Shakespeare at school Picasso is on the far wall My Rolling Stones is missing Brazil wins World Cup. China wins first round of trade talks ‘There is a pragmatic connection between proper names and objects that will be conventionally associated, within a culturally defined community, with those names Using a proper name referentially to identify such an object invites the listener to make the expected inference (from name of writer to book by writer) and thus show to be a member of the same community as the speaker. In such cases, it becomes obvious that ‘more is being communicated than is said. The role of co-text Tin istic material, or co-text, limits the range of possible interpretations we may have for any oF the above examples. it is misleading to think that reference is understood solely in terms of our ability to identify referents via the referring expression, ‘The physical environment, or context, has a powerful impact on how referring ‘expressions are to be interpreted. For example, it is useful to know that a hospital, a dentist's practice, and a hotel reception are the contexts for ‘The Hrt-attack mustn’t be moved. ‘Your two-thirty just cancelled, A couple of rooms have complained about the heat. Reference is not simply a relationship between the meaning of a word or phrase and an ‘object or person in the world. It is a social act in which the speaker assumes that the ‘word or phrase chosen to identify an object or person are interpreted as the speaker intended, Entailment Entailments follow from the sentence, regardless of the speaker's beliefs. They are communicated without being sada venience A entalsTsectonee 8 fier iuek ofB follows logically from the truth of A. For example: Mary ate all the apples. entails: ‘Someone ate something. Entailments are divided into background and foreground based on the principle of informativeness (Wilson and Sperber 1979) and on the grammatical form, including stress. In the example: John is married to Ann. the stress determines a chain of entailments and also the focal scale: John is married to someone. John has some property Something is the case ‘These are entailments obtained by substitution. Atlas and Levinson (1981) emphasized the role of logical form, of the structure of semantic Tepresentation, in the production of pragmatic inferences, in identifying what the sentence is about, Consider{twas John that Mary kissed. (x) (y) (a group kissed by Mary has the property of being identical to John) ‘The logical subject is thus “A group kissed by Mary’—and this is what the sentence is. about; this corresponds to the surface structure ‘one that Mary kissed’ which entails that Mary kissed someone, and that Mary kissed John, but it has not the same truth conditions as the unclefted “Mary kissed John’ since it entails that Mary kissed just John, Because of its logical nature, entailment is not much discussed in contemporary pragmatics as the speaker-dependent notion of presupposition PRESUPPOSITION ‘This type of pragmatic inference differs from conversational implicature (hased On conte\tuz! aeswmptions conceming the co-orerctiveness of participants in a conversation) in that it is based more closely on the actual ling istic structure of ntences (not semantic in the narrow sense because they are very sensitive to contextual tors). Pragmatic presuppositions are relations between the speaker and the appropriateness ofa sentence ina comes If we say that the sentence a contains the Proposition p and the sentence b Contains the proposition q, then, using the symbol >> to mean ‘presupposes’ we can ¢ represent the relationship as follows: a Mary’s dog is cute, (~p) b. B.Mary has adog. (~g) eee: When we produce the opposite of the sentence in a by negating it (-NOT), we find that the relationship of presupposition doesn’t change: a. Mary’s dog isn’t cute. (=NOT p) . b. B. Mary has adog. (=g) ©. NOT p>>q This property of presupposition is generally described as constancy under negation, ‘meaning that the presupposition of a statement remains constant even when That statement is negated. G. Frege’s and P. F. Strawson’s claim that Presuppositions are preserved in negative sentences or statements, provides an initial test for identifying presuppositions We can simply take a sentence, negate it and See what inference sarrce eos the by both the positive and negative sentence. John managed to stop in time. From this we may infer: ‘John stopped in time’ and ‘John tried to ‘stop in time.” Now consider the sentence: John didn’t manage to stop in time. From this we cannot infer “John stopped in time” yet the inference to “John tried to stop in time’ is preserved and shared by both the positive and negative sentences. The result isa presupposition for both positive and negative. ‘The presupposition comes from “manage’—presuppositions tend to be tied to particular words (or other aspects of theSurface structure). Presupposition zenerating linguistic items are called presuppasition- ~Tonsider: John, who is a good friend of mine. regrets that he stopped doing linguistics before he left Sibiu And the negative: John, who is a good friend of mine, doesn’t regret that he stopped doing linguisties before he left Sibiu The following set of inferences can be derived from these sentences: -there's someone identifiable to speaker and addressee as John ~John is a good friend of the speaker's ~John stopped doing linguistics before he left Sibiu ~John was doing linguistics before he lett Sibiu John left Sibiu Sine these are invariable under nevation, they candidate as presuppositions under Frege/Strawson conception, They can be tied to specitic words and constructions that ‘generate them, Sources of presuppositions: through constructing a list of presupposition triggers. Selection: 1. Definite descriptions John saw/didn’t see the man with two heads. (>There exists a man with two heads) 2. Factive verbs Martha regrets/doesn’t regret drinking John's wine. (M. drank it) She didn’t realize he was ill. (>>He was ill) We regret telling him, (>>We told him) It isn’t odd that he left early. (>>He left early) Other verbs that have factive presuppositions are: realize, be aware, know, be sorry that, regret, be proud that, be indifferent that, be glad that, be sad that 3. Implicative verbs (manage, forget, happen to, not plan to, not intend to, avoid, be expected to, ought to) John managed/didn’t manage to open the door (J. tried) Change of state verbs that generate lexical presupposition: stop, star, finish, carry on, cease, leave, enter, come, go, arrive. John stoppedididn’t stop beating his wife (>>John had been beating his wife) Joan began/didn’t begin to beat her husband (Joan>> hadn't been beating her husband) Clinton continued/didn’t continue ruling the world (>>Clinton had been ruling the world) He stopped smoking, (>>He used to smoke) 5. Iteratives such as: again, return, another time, to come back, restore, repeat, for the nth time, ‘The flying saucer came again. (>>It came before) Iliescu returned to power. (>>lliescu held power before) You're late again. (>>You were late before) 6. Verbs of judging (accuse, criticize)‘Ann accusedididn’t accuse John of plagiarism (>>Ann thinks plagiarism is bad.) 7. Temporal clauses introduced by: before, after, during, whenever. as Pefore Sandy was born, his parents livediidn’t live in another place. (>>Sandy was bem) While Chomsky was revolutionizing linguistics, the rest of the social sciences was/wasn't asleep. (>>Chomsky revolutionized linguistics) Since Churchill died, English have lacked/haven't lacked a leader. (>>Churchill died) 8. Cleft sentences, tha is a sentence which has been divided into two parts, each with its pun verb, to emphasize a particular piece of information. Cleft sentences usually besin with “it plus a form ofthe verb “be' followed by the element which is being emphasized. Pseudo-cleft sentences begin with wh- It was/wasn’t John that kissed Anne. (>>Someone did) iefted: John kissed Mary. (>>Someone dia) What John lost/didn’t lose was his wallet (>>John lost something) ‘What | need is a good holiday. (>>f need something) or: A good holiday is what I need. 9. Comparisons and contrasts such as: to, back, in return, or by comparative constructions Mary called Adam a male chauvinist, and then he insulted her (in return). (>>Male chauvinist is an insult) 10. Counterfactual conditionals meaning that what is presupposed is not only not true, but is the opposite of what is true, or ‘contrary to facts.” A conditional structure of this type presupposes that the information in the if-clause is not true at the time of the utterance. if the plane had landed on time Nick and Susan would have met. (>>The plane had not landed on time) Ifyou were my friend, you would have helped me. (>>You are not my friend) 11. Yes/No questions 1s there a professor of linguistics at this university? (>>Either there is or there isn’t) All these categories depend crucially on the definitions of presupposition. Constancy ‘under negation is often not enough to establish a coherent set of inferences, Presuppositions have the following properties: 2) they are defeasible, which proves that any semantic theory of presupposition can be undone ») they are apparently tied to particular aspects of surface structure, which distinguishes Presupposition from conversational implicatures. Presuppositions are not stable, context-independent aspects of meaning and therefore they belong to pragmatics. The projection problem ‘There is a basic expectation that the presupposition of a simple sentenc ontinue to bbe true when that simple sentence becomes part of a more complex sentence This is oneversion of the general idea that the meaning of the whale sentence is a combination of the meaning of ts parts. However, the meaning of some presuppositions doesn't sur become the meaning of same complex sentences, This is known as the pro} problem. Consider: —— = a. Nobody realized that Kelly was ill. (=p) Kelly was il. (=4) p>>q (At this point, the speaker uttering a, presupposes b.) imagined that Kelly was ill, (=r) Kelly was not ill, (NOT 4) FeNOT q (At thisd point, the speaker utering d. presupposes e. the opposite ~ ofb,) a = {imagined that Kelly was ill and nobody realized that she avs ill. (=rp) x r&pNOT g (At ts point. after combining r & p, the presunposition g can no longer be assumed to be true) The presupposition q (Kelly was ill) is assumed to be true in the simple structure but doesn’t project into the complex structure. In this example, the technical analysis is logical. but itis difficult to think of a context in which someone would talk like that Pragmatic theories of presuppositions I. Developed by Karttunen and Peters (1979), expressed in the framework of Montagué Srammar. in which clauses are built from the bottom up rather than from the top down 4s in transformational generative grammar. Thus every word, clause or syntactic operation can have associated with it a semantic representation OT extension expression ‘They add to the Framework of Montag ‘ditional Set oF meaning expressions to be generated in the same sort as extension expressions, as sentences are built up from their constituent parts—associated with presupposition-iriggers Karttunen and Peters connect their theory to the earlier attempts to define pragmatic Presupposition, along the following lines: cooperative participants have the obligation to “organize their contributions in such a way that the conventional implicature of the sentence uttered are already part of the common ground at the time of the utterance” (1975). There are some problems with this theory because the solutions 1 suggests are simply of an engineering sort and cannot handle contextual defeasibilty 2. Gazdar’s theory, which basically asserts that the order in which an utterance’s inferences are added is: entailments of the utered sentence, conversational implicatures, presuppositions. For example Ifthere is a king of France, the king of France doesn’t any longer live in Versailles. ‘The speaker knows that there exists a king of France. It is consistenf-with all the speaker knows that there is not a king of France. Both theories assume that each presupposition-trigger has its own presupposition recorded in the lexicon or elsewhere in the surface structure. A more comprising theory should be able to predict presuppositions from the semantic Content of presupposition-triggers, by means of a general pragmatic principle but. ‘unfortunately, presupposition remains only partially understood and more research is required of how semantics and pragmatics interact in order to further such studies, Ss