DOCSLIB.ORG

Are Natural Language Inference Models Imppressive? Learning Implicature and Presupposition

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Are Natural Language Inference Models Imppressive? Learning Implicature and Presupposition Are Natural Language Inference Models IMPPRESsive? Learning IMPlicature and PRESupposition Paloma Jereticˇ∗1, Alex Warstadt∗1, Suvrat Bhooshan2, Adina Williams2 1Department of Linguistics, New York University 2Facebook AI Research fpaloma,[email protected], fsbh,[email protected] Abstract Natural language inference (NLI) is an in- creasingly important task for natural language understanding, which requires one to infer whether a sentence entails another. However, the ability of NLI models to make pragmatic inferences remains understudied. We create an IMPlicature and PRESupposition diagnostic dataset (IMPPRES), consisting of >25k semi- automatically generated sentence pairs illus- trating well-studied pragmatic inference types. Figure 1: Illustration of key properties of classical en- We use IMPPRES to evaluate whether BERT, tailments, implicatures, and presuppositions. Solid ar- InferSent, and BOW NLI models trained on rows indicate valid commonsense entailments, and ar- MultiNLI (Williams et al., 2018) learn to make rows with X’s indicate lack of entailment. Dashed ar- pragmatic inferences. Although MultiNLI rows indicate follow up statements with the addition of appears to contain very few pairs illustrat- in fact, which can either be acceptable (marked with ing these inference types, we find that BERT ‘7’) or unacceptable (marked with ‘3’). learns to draw pragmatic inferences. It re- liably treats scalar implicatures triggered by “some” as entailments. For some presuppo- are separate semantic and pragmatic modes of rea- sition triggers like only, BERT reliably recog- nizes the presupposition as an entailment, even soning (Grice, 1975; Clark, 1996; Beaver, 1997; when the trigger is embedded under an entail- Horn and Ward, 2004; Potts, 2015), and it is not ment canceling operator like negation. BOW clear which of these modes, if either, NLI mod- and InferSent show weaker evidence of prag- els learn. We investigate two pragmatic inference matic reasoning. We conclude that NLI train- types that are known to differ from classical en- ing encourages models to learn some, but not tailment: scalar implicatures and presuppositions. all, pragmatic inferences. As shown in Figure1, implicatures differ from en- 1 Introduction tailments in that they can be denied, and presuppo- sitions differ from entailments in that they are not One of the most foundational semantic discover- canceled when placed in entailment-cancelling en- ies is that systematic rules govern the inferential vironments (e.g., negation, questions). relationships between pairs of natural language To enable research into the relationship be- sentences (Aristotle, De Interpretatione, Ch. 6). tween NLI and pragmatic reasoning, we introduce In natural language processing, Natural Language IMPPRES, a fine-grained NLI-style diagnostic test Inference (NLI)—a task whereby a system de- dataset for probing how well NLI models perform termines whether a pair of sentences instantiates implicature and presupposition. Containing 25.5K in an entailment, a contradiction, or a neutral sentence pairs illustrating key properties of these relation—has been useful for training and evaluat- pragmatic inference types, IMPPRES is automati- ing models on sentential reasoning. However, lin- cally generated according to linguist-crafted tem- guists and philosophers now recognize that there plates, allowing us to create a large, lexically var- ∗Equal Contribution ied, and well controlled dataset targeting specific 8690 Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 8690–8705 July 5 - 10, 2020. c 2020 Association for Computational Linguistics instances of both types. Type Example We first investigate whether presuppositions Trigger Jo’s cat yawned. and implicatures are present in NLI models’ train- Presupposition Jo has a cat. ing data. We take MultiNLI (Williams et al., Negated Trigger Jo’s cat didn’t yawn. 2018) as a case study, and find it has few in- Modal Trigger It’s possible that Jo’s cat yawned. Interrog. Trigger Did Jo’s cat yawn? stances of pragmatic inference, and almost none Cond. Trigger If Jo’s cat yawned, it’s OK. that arise from specific lexical triggers (see x4). Negated Prsp. Jo doesn’t have a cat. Given this, we ask whether training on MultiNLI Neutral Prsp. Amy has a cat. is sufficient for models to generalize about these largely absent commonsense reasoning types. We Table 1: Sample generated presupposition paradigm. find that generalization is possible: the BERT Examples adapted from the ‘change-of-state’ dataset. NLI model shows evidence of pragmatic reason- ing when tested on the implicature from some to of implicatures we focus here on scalar implica- not all, and the presuppositions of certain triggers tures. Scalar implicatures are inferences, often (only, cleft existence, possessive existence, ques- optional,2 which can be drawn when one mem- tions). We obtain some negative results, that sug- ber of a memorized lexical scale (e.g., hsome, alli) gest that models like BERT still lack a sophisti- is uttered (see x6.1). For example, when some- cated enough understanding of the meanings of the one utters Jo ate some of the cake, they suggest lexical triggers for implicature and presupposition that Jo didn’t eat all of the cake, (see Figure1 (e.g., BERT treats several word pairs as synonyms, for more examples). According to Neo-Gricean e.g., most notably, or and and). pragmatic theory (Horn, 1989; Levinson, 2000), Our contributions are: (i) we provide a new the inference Jo didn’t eat all of the cake arises diagnostic test set to probe for pragmatic infer- because some has a more informative lexical al- ences, complete with linguistic controls, (ii) to our ternative all that could have been uttered instead. knowledge, we present the first work evaluating We expect the speaker to make the most informa- deep NLI models on specific pragmatic inferences, tive true statement:3 as a result, the listener should and (iii) we show that BERT models can perform infer that a stronger statement, where some is re- some types of pragmatic reasoning very well, even placed by all, is false. when trained on NLI data containing very few ex- Implicatures differ from entailments (and, as we plicit examples of pragmatic reasoning. We pub- will see, presuppositions; see Figure1) in that they licly release all IMPPRES data, models evaluated, are deniable, i.e., they can be explicitly negated annotations of MultiNLI, and the scripts used to without resulting in a contradiction. For example, 1 process data. someone can utter Jo ate some of the cake, fol- lowed by In fact, Jo ate all of it. In this case, the 2 Background: Pragmatic Inference implicature (i.e., Jo didn’t eat all the cake from We take pragmatic inference to be a relation be- above) has been denied. We thus distinguish im- tween two sentences relying on the utterance con- plicated meaning from literal, or logical, meaning. text and the conversational goals of interlocu- 2.2 Presupposition tors. Pragmatic inference contrasts with seman- tic entailment, which instead captures the logical Presuppositions of a sentence are facts that the relationship between isolated sentence meanings speaker takes for granted when uttering a sentence (Grice, 1975; Stalnaker, 1974). We present impli- (Stalnaker, 1974; Beaver, 1997). Presuppositions cature and presupposition inferences below. are generally associated with the presence of cer- tain expressions, known as presupposition trig- 2.1 Implicature gers. For example, in Figure1, the definite de- Broadly speaking, implicatures contrast with en- 2Implicature computation can depend on the cooperativity tailments in that they are inferences suggested by of the speakers, or on any aspect of the context of utterance the speaker’s utterance, but not included in its lit- (lexical, syntactic, semantic/pragmatic, discourse). See De- gen(2015) for a study of the high variability of implicature eral (Grice, 1975). Although there are many types computation, and the factors responsible for it. 3This follows if we assume that speakers are cooperative 1github.com/facebookresearch/ImpPres (Grice, 1975) and knowledgeable (Gazdar, 1979). 8691 scription the cake triggers the presupposition that NLI itself has been steadily gaining in popular- there is a cake (Russell, 1905). Other examples of ity; many datasets for training and/or testing sys- presupposition triggers are shown in Table1. tems are now available including: FraCaS (Cooper Presuppositions differ from other inference et al., 1994), RTE (Dagan et al., 2006; Mirkin types in that they generally project out of opera- et al., 2009; Dagan et al., 2013), Sentences In- tors like questions and negation, meaning that they volving Compositional Knowledge (Marelli et al., remain valid inferences even when embedded un- 2014, SICK), large scale imaging captioning as der these operators (Karttunen, 1973). The infer- NLI (Bowman et al., 2015, SNLI), recasting ence that there is a cake survives even when the other datasets into NLI (Glickman, 2006; White presupposition trigger is in a question (Did Jor- et al., 2017; Poliak et al., 2018), ordinal common- dan eat some of the cake?), as shown in Figure1. sense inference (Zhang et al., 2017, JOCI), Multi- However, in questions, classical entailments and Premise Entailment (Lai et al., 2017, MPE), NLI implicatures disappear. Table1 provides exam- over multiple genres of written and spoken En- ples of triggers projecting out of several entail- glish (Williams et al., 2018, MultiNLI),
Load more

Recommended publications
  • Making Inferences: the Case of Scalar Implicature Processing Judith Degen ([email protected]) Michael K
    Making Inferences: The Case of Scalar Implicature Processing Judith Degen ([email protected]) Michael K. Tanenhaus ([email protected]) Department of Brain and Cognitive Sciences, University of Rochester Rochester, NY 14627-0268, USA Abstract ten). Under the assumption that a speaker is being as informa- tive as possible, a weak implicature, that the speaker doesn’t Scalar implicature has served as a test case for investigating the nature of inference processes in language comprehension. know whether all of the sections are written, is licensed. If Specifically, the question of whether or not scalar implicatures the hearer further assumes that the speaker is competent with are computed by default has been extensively investigated in respect to the truth of the stronger statement, the implicature recent years. We argue that the question of default is overly simplistic and propose instead to think of scalar implicature that the speaker believes that some, but not all of the sections computation as a problem of optimal cue combination within a are written, is licensed. constraint-based framework. We provide evidence from three In recent years the representation and processing of scalar experiments supporting the view that multiple constraints of differing strength operate in parallel to provide probabilistic implicatures has emerged as perhaps the defining problem in support for or against an implicature. experimental pragmatics – a subfield of cognitive science that Keywords: experimental pragmatics; scalar implicature; eye- seeks to combine theoretical proposals from linguistics, com- tracking; subitizing puter science and philosophy with state-of-the-art experimen- tal methods. Introduction Importantly for the experimental investigation of the phe- Successful communication requires comprehenders to infer a nomenon, scalar implicatures are cancelable, that is, they are speaker’s intended meaning from an underspecified utterance.
    [Show full text]
  • Grice's Theory of Implicature and the Meaning of the Sentential
    Gergely Bottyán The operationality of Grice’s tests for implicature 1. Introduction In less than a decade after its first publication in 1975, Herbert Paul Grice’s paper Logic and conversation becomes one of the classic treatises of the linguistic subdiscipline now standardly referred to as pragmatics. There are at least two reasons for the paper’s success: (i) it can be regarded as the first truly serious attempt to clarify the intuitive difference between what is expressed literally in a sentence and what is merely suggested by an utterance of the same string of words, (ii) the components of the notional and inferential framework that Grice set up to characterize various kinds of utterance content are intuitively appealing (cf. Haberland & Mey 2002). The present paper has two related aims: (i) to give a comprehensive overview of Grice’s theory of implicature, and (ii) to investigate whether it is possible to test the presence of a conversational implicature on the basis of some or all of the properties that Grice attributes to this construct. It will be argued that of the six characteristics listed by Grice, only cancellability is a practical criterion. However, cancellability is not a sufficient condition of the presence of a conversational implicature. Therefore, additional criteria are needed to arrive at an adequate test. 2. Overview of Grice’s theory of implicature Most of the body of Grice 1975=1989a consists in an attempt to clarify the intuitive difference between what is expressed literally in a sentence and what is merely suggested or hinted at by an utterance of the same string of words.
    [Show full text]
  • Conditional Strengthening [A Case Study in Implicature]
    Draft • Version from 9/13/01 1:08 PM Conditional Strengthening [A Case Study in Implicature] Kai von Fintel MIT Introduction Hearers can extract from an asserted sentence a good amount of information that goes beyond the grammatically encoded truth-conditions of the sentence. There are inferences of at least two kinds: (i) Inferences from the (assumed) truth of the asserted sentence. Hearers may have conditional beliefs (if p, q) and upon hearing p asserted they can infer q by Modus Ponens (with suitable caveats about the reliability of their initial conditional belief and the new information that p). (ii) Inferences from the fact that the sentence was asserted. Hearers may have conditional beliefs of the following nature: if q were the case, p would not have been asserted. This together with the information that p was in fact asserted leads via Modus Tollens to the conclusion that q cannot have been the case. For both of these kinds of inferences, we can ask whether the speaker bears (some) responsibility for them. If the speaker is (or should be) well aware of the relevant conditional beliefs on the part of the hearer, he or she may be partly responsible for the inference. Many authors argue that such inferences are part of the utterance meaning if the underlying assumptions are mutual beliefs of speaker and hearer. It may often be much easier to decide whether speaker bears some responsibility for an inference than to decide whether the inference is part of the meaning of the utterance. I will not worry about these issues of responsibility.
    [Show full text]
  • 1. Introduction
    MICHELA IPPOLITO PRESUPPOSITIONS AND IMPLICATURES IN COUNTERFACTUALS In this article, I propose a pragmatic account of temporally mismatched subjunctive past counterfactuals. The proposal consists of the following parts. First, I show that in cases of temporal mismatch, Past cannot be interpreted inside the proposition where it occurs at surface structure. Instead, it must be interpreted as constraining the time variable of the accessibility relation. This has the effect of shifting the time of the evaluation of the conditional to some contextually salient past time. Second, I will propose specific Felicity Conditions (presuppositions) for subjunctive conditionals and I will argue that there is a strict correspondence between the time of evaluation in the truth-conditions of a conditional and the time relevant for the felicity conditions. In other words, if the time relevant for the accessibility relation has been shifted to the past, then the conditional’s presupposition will make reference to the speaker’s past epistemic states. On the other hand, if no past is constraining the time argument of the accessibility relation, the conditional’s felicity will make reference to the speaker’s epistemic state at the time of the utterance. Third, I will argue that the intuition that the antecedent of mismatched counterfactuals is not true is a scalar implicature arising not from a competition between assertions but between presuppositions. Finally, I will investigate the general repercussions of my proposal on the problem of presupposition projection. 1. Introduction Past subjunctive conditionals do not always talk about the past. Sometimes they may talk about the future, despite the overt past verbal morphology.
    [Show full text]
  • The Acquisition of Scalar Implicatures
    Wer bei „scalar implicatures“ an eine exotische Zierfischart denkt und vor seinem geistigen Auge bunte Segelflosser majestätisch zwischen Steinen und Grünpflanzen dahinschweben sieht, der irrt sich sehr: Stefanie Röhrig Scalar implicatures bezeichnen keine Buntbarschart, sondern gehören in den Bereich der Pragmatik. Diese Disziplin der Linguistik untersucht sprachliches Handeln, also sprachliche Äußerungen und die Entstehung von Bedeutung. Insbesondere logische Schlussfolgerungen sind von wissenschaftlichem The Acquisition of Interesse. Wer auf die Frage „Have you read Shakespeare‘s works?“ mit „I have read some of them“ antwortet, der bejaht einerseits die Frage, schränkt aber - ohne dies ausdrücklich zu sagen - zugleich die Zustimmung ein: some of them heißt nicht all of them. Implikaturen müssen ‚zwischen den Zeilen‘ Scalar Implicatures verstanden werden. Gewöhnlich wird davon ausgegangen, dass Kinder erst mit etwa 6 Jahren Implikaturen wirklich verstehen und für sich nutzen können. Stefanie Röhrig zeigt in ihrer als Feldstudie angelegten Göttinger Schriften zur Englischen Philologie Staatsarbeit, dass diese Lehrmeinung dringend der Überarbeitung bedarf. Denn Kinder, so decken ihre Untersuchungen auf, verfügen sogar über eine ihnen eigene Art der Implikatur. Band 3 2010 Stefanie Röhrig The Acquisition of Scalar Implicatures Scalar of Acquisition The Röhrig Stefanie ISBN: 978-3-941875-49-4 ISSN: 1868-3878 Universitätsdrucke Göttingen Universitätsdrucke Göttingen Stefanie Röhrig The Acquisition of Scalar Implicatures This work is licensed under the Creative Commons License 3.0 “by-nd”, allowing you to download, distribute and print the document in a few copies for private or educational use, given that the document stays unchanged and the creator is mentioned. You are not allowed to sell copies of the free version.
    [Show full text]
  • Conversational Implicatures (And How to Spot Them)
    Conversational Implicatures (and How to Spot Them) Michael Blome-Tillmann McGill University Abstract In everyday conversations we often convey information that goes above and beyond what we strictly speaking say: exaggeration and irony are obvious examples. H.P. Grice intro- duced the technical notion of a conversational implicature in systematizing the phenom- enon of meaning one thing by saying something else. In introducing the notion, Grice drew a line between what is said, which he understood as being closely related to the conventional meaning of the words uttered, and what is conversationally implicated, which can be inferred from the fact that an utterance has been made in context. Since Grice’s seminal work, conversational implicatures have become one of the major re- search areas in pragmatics. This article introduces the notion of a conversational implica- ture, discusses some of the key issues that lie at the heart of the recent debate, and expli- cates tests that allow us to reliably distinguish between semantic entailments and conven- tional implicatures on the one hand and conversational implicatures on the other.1 Conversational Implicatures and their Cancellability In everyday conversations our utterances often convey information that goes above and beyond the contents that we explicitly assert or—as Grice puts it—beyond what is said by our utterances. Consider the following examples:2 (1) A: Can I get petrol somewhere around here? B: There’s a garage around the corner. [A can get petrol at the garage around the corner.] (2) A: Is Karl a good philosopher? B: He’s got a beautiful handwriting.
    [Show full text]
  • The Phrasal Implicature Theory of Metaphors and Slurs
    THE PHRASAL IMPLICATURE THEORY OF METAPHORS AND SLURS Alper Yavuz A Thesis Submitted for the Degree of PhD at the University of St Andrews 2018 Full metadata for this thesis is available in St Andrews Research Repository at: http://research-repository.st-andrews.ac.uk/ Please use this identifier to cite or link to this thesis: http://hdl.handle.net/10023/13189 This item is protected by original copyright This item is licensed under a Creative Commons Licence The Phrasal Implicature Theory of Metaphors and Slurs Alper Yavuz This thesis is submitted in partial fulfilment for the degree of Doctor of Philosophy (PhD) at the University of St Andrews March 2018 Abstract This thesis develops a pragmatic theory of metaphors and slurs. In the pragmatic literature, theorists mostly hold the view that the framework developed by Grice is only applicable to the sentence-level pragmatic phenomena, whereas the subsenten- tial pragmatic phenomena require a different approach. In this thesis, I argue against this view and claim that the Gricean framework, after some plausible revisions, can explain subsentential pragmatic phenomena, such as metaphors and slurs. In the first chapter, I introduce three basic theses I will defend and give an outline of the argument I will develop. The second chapter discusses three claims on metaphor that are widely discussed in the literature. There I state my aim to present a theory of metaphor which can accommodate these three claims. Chapter3 introduces the notion of \phrasal implicature", which will be used to explain phrase- level pragmatic phenomena with a Gricean approach. In Chapter4, I present my theory of metaphor, which I call \phrasal implicature theory of metaphor" and discuss certain aspects of the theory.
    [Show full text]
  • Bare Numerals and Scalar Implicatures
    Language and Linguistics Compass 7/5 (2013): 273–294, 10.1111/lnc3.12018 Bare Numerals and Scalar Implicatures Benjamin Spector* CNRS-ENS-EHESS, Institut Jean Nicod, École Normale Supérieure Abstract Bare numerals present an interesting challenge to formal semantics and pragmatics: they seem to be compatible between various readings (‘at least’, ‘exactly’, and ‘at most’ readings), and the choice of a particular reading seems to depend on complex interactions between contextual factors and linguistic structure. The goal of this article is to present and discuss some of the current approaches to the interpretation of bare numerals in formal semantics and pragmatics. It discusses four approaches to the interpretation of bare numerals, which can be summarized as follows: 1. In the neo-Gricean view, the basic, literal meaning of numerals amounts to an ‘at least interpreta- tion’, and the ‘exactly n’ reading results from a pragmatic enrichment of the literal reading, i.e. it is accounted for in terms of scalar implicatures. 2. In the underspecification view, the interpretation of numerals is ‘underspecified’, with the result that they can freely receive the ‘at least’, ‘exactly’,or‘at most’ reading, depending on which of these three construals is contextually the most relevant. 3. In the ‘exactly’-only view, the numerals’ basic, literal meaning corresponds to the ‘exactly’ reading, and the ‘at least’ and ‘fewer than n’ readings result from the interaction of this literal meaning with background, non-linguistic knowledge. 4. In the ambiguity view, numerals are ambiguous between two readings, the ‘at least’ and ‘exactly’ readings.The article argues that in order to account for all the relevant data, one needs to adopt a certain version of the ambiguity view.
    [Show full text]
  • Relevance Implicatures
    Relevance Implicatures Robert van Rooy∗ ILLC/University of Amsterdam [email protected] Abstract According to standard pragmatics, we should account for conversational implicatures in terms of Grice’s (1967) maxims of conversation. Neo-Griceans like Atlas & Levinson (1981) and Horn (1984) seek to reduce those maxims to the so-called Q and I-principles. In this paper I want to argue that (i) there are major problems for reducing Gricean pragmatics to these two principles, and (ii) in fact, we can better account for implicatures by means of an exhaustivity operator defined in terms of a relevance-based ordering relation. 1 Introduction In most circumstances we infer from the fact that the (a) sentences of the following examples are uttered that the (b) sentences are true: (1) a. John ate some of the cookies. b. John didn’t eat all of the cookies (2) a. John is coming or Mary is going. b. John is coming or Mary is going, but not both. (3) a. John is walking, if Mary is talking. b. John is walking if and only if Mary is talking. ∗This note reports on work in progress on the use of relevance for the semantic analysis of natural language. Parts of this paper was presented at the second workshop on Optimality Theoretic Semantics at OTS in Utrecht (2000) and at the Formal Pragmatics conference in Berlin (2001). I would like to thank the participants of those events for their comments and suggestions. I would also like to thank the anonymous reviewers, Reinhard Blutner and expecially Katrin Schulz.
    [Show full text]
  • Scalar Implicatures Vs. Presuppositions – the View from Acquisition
    Scalar Implicatures vs. Presuppositions – The view from Acquisition Cory Bill, Jacopo Romoli, Florian Schwarz, and Stephen Crain To appear in TOPOI, special issue “Presuppositions: Philosophy, Linguistics, and Psychology.” This paper reports an experimental investigation of presuppositions and scalar implicatures in language acquisition. Recent proposals (Chemla 2009; Romoli 2012, 2014) posit the same mechanisms for generating both types of inferences, in contrast to the traditional view. We used a Covered Box picture selection task to compare the interpretations assigned by two groups of children (4/5 and 7 year olds) and by adults, in response to sentences with presuppositions and ones with either ‘direct’ or ‘indirect’ scalar implicatures. The main finding was that the behavior of children and adults differed across inference types. This asymmetry is consistent with the traditional perspective, but poses a challenge for the more recent uniform accounts. We discuss how the latter could be amended to account for these findings, and also relate the findings to previous results on presupposition processing. Finally, we discuss an unexpected difference found between direct and indirect scalar implicatures. Keywords: Presuppositions, Scalar Implicatures, Pragmatics, Child Language Acquisition. 1 Introduction As the sentences in (1a)-(3a) illustrate, scalar implicatures like those in (1b) and (2b), and presuppositions such as (3b) are central features of the overall meaning that is conveyed when speakers make certain utterances. (1) a. Some of the giraffes have scarves b. Not all of the giraffes have scarves (2) a. Not all of the giraffes have scarves b. Some of the giraffes have scarves (3) a. The bear didn’t win the race b.
    [Show full text]
  • Discontinuous Past’: the Curious Implicatures of Optional Past Tense 1
    The Tlingit Decessive and ‘Discontinuous Past’: The Curious Implicatures of Optional Past Tense 1 Seth Cable University of Massachusetts Amherst ABSTRACT: Some languages appear to have a morpheme that combines the meaning of past tense with a variety of additional implications, the nature of which depend upon the aspectual marking of the verb. For non-perfective verbs (imperfective, habitual, future, etc.), the additional implication is that the event/state in question fails to extend into the present. For perfective verbs, however, the additional implication is either that (i) the result state of the event fails to extend into the present, or (ii) some natural, expected consequence of the event failed to occur. Prior authors have viewed these additional inferences as semantic in nature, as being encoded directly in the lexical semantics of the morpheme (Leer 1991; Copley 2005; Plungian & van der Auwera 2006). Under this view, the morphemes in question express a special category of tense, one that has been labeled ‘discontinuous past’ by Plungian & van der Auwera (2006). Through in-depth investigation of one such ‘discontinuous past’ marker in the Tlingit language, I argue that – to the contrary – the special inferences of these morphemes are not semantic, and are instead defeasible pragmatic inferences. Consequently, putative instances of ‘discontinuous past’ are in their semantics simply past tenses. I provide a formalized analysis of the pragmatic inferences associated with these past tenses, whereby they ultimately follow from (i) the optionality of the tense markers in question, and (ii) a special principle relating to the inherent topicality of the utterance time. The empirical and analytic results align well with a restrictive theory of cross- linguistic variation in tense semantics, one where the ‘traditional’ categories of PAST, PRESENT, and FUTURE are the only ones representable in human language (Cable 2013).
    [Show full text]
  • Types of Implicature in Informal Conversations Used by the English Education Study Program Students
    Rahayu, Types Of Implicature In Informal… TYPES OF IMPLICATURE IN INFORMAL CONVERSATIONS USED BY THE ENGLISH EDUCATION STUDY PROGRAM STUDENTS Endry Sri Rahayu Prof.Safnil M.A,Ph.D. University of Bengkulu W.R. Supratman, Kd. Limun 38371A. Bengkulu, Indonesia Email : [email protected] Email : [email protected] Abstract: Implicature was refers to implied meaning in utterance that can be understood by indirectly expression. In informal conversation was occurred the hidden meaning of what actually said by the speakers. This study was investigated the types of implicature in informal conversations used by the English education study program students. The study was aimed to analyze the types of implicature and how the implicature is carried out in the informal conversations. The method of study was a descriptive qualitative method. The subjects of this study were 25 students of English study program who have informal conversation. The students’ conversation was transcribed and analyzed by using checklist instrument. The results was shown that 1) there were three types of implicature found in the informal conversations; conventional implicature, generalized conversational implicature and particularized conversational implicature, and 2) the implicature is carried out in the informal conversations by the used of generalized conversational implicature and particularized conversational implicature. Moreover, a conclusion is students in the informal conversation have potentially implicature that indicates that their utterance has implied meaning. The suggestions, the study about implicature should be conducted in different area such as movie, drama or others, to give enrichment information in the Pragmatics study. Keywords: Implicature Types, Informal Conversations, English Education Study Program Students. Abstrak: Implikatur mengacu pada makna tersirat pada suatu ujaran yang dapat dipahami dari ekspresi tidak langsung.
    [Show full text]