Experience-grounded semantics: a theory for intelligent systems

An experience-grounded semantics is introduced for an intelligent reasoning system, which is adaptive, and works with insufficient knowledge and resources. According to this semantics, truth and meaning are defined with respect to the experience of the system – the truth value of a statement indicates the amount of available evidence, and the meaning of a term indicates its experienced relations with other terms. The major difference between experience-grounded semantics and model-theoretic semantics is that the former does not assume the sufficiency of knowledge and resources. This approach provides new ideas to the solution of some important problems in cognitive science.     

Game theoretical semantics and entailment

The essence of the meaning of a declarative sentence is given by stating its truth conditions, and consequently semantics, the study of meaning, must include a theory of truth conditions. Such a theory must not only describe accurately the truth conditions of declarative sentences, it must also answer the question of when two sentences have the same truth conditions. The fundamental semantic relation of having the same truth conditions cannot be ignored by any reasonable theory.This paper is an attempt to find a partial account of this relation by using game theoretical semantics as developed by Hintikka and his followers. The account given will establish a connection between this approach to semantics and the theory of firstdegree entailment formulated by Anderson and Belnap.     

One's Modus Ponens: Modality,Coherence and Logic

Recently, there has been a shift away from traditional truth‐conditional accounts of meaning towards non‐truth‐conditional ones, e.g., expressivism, relativism and certain forms of dynamic semantics. Fueling this trend is some puzzling behavior of modal discourse. One particularly surprising manifestation of such behavior is the alleged failure of some of the most entrenched classical rules of inference; viz., modus ponens and modus tollens. These revisionary, non‐truth‐conditional accounts tout these failures, and the alleged tension between the behavior of modal vocabulary and classical logic, as data in support of their departure from tradition, since the revisionary semantics invalidate some of these patterns. I, instead, offer a semantics for modality with the resources to accommodate the puzzling data while preserving classical logic, thus affirming the tradition that modals express ordinary truth‐conditional content. My account shows that the real lesson of the apparent counterexamples is not the one the critics draw, but rather one they missed: namely, that there are linguistic mechanisms, reflected in the logical form, that affect the interpretation of modal language in a context in a systematic and precise way, which have to be captured by any adequate semantic account of the interaction between discourse context and modal vocabulary. The semantic theory I develop specifies these mechanisms and captures precisely how they affect the interpretation of modals in a context, and do so in a way that both explains the appearance of the putative counterexamples and preserves classical logic.     

Lepore and Ludwig on ‘explicit meaning theories’

The fundamental problem proponents of truth conditional semantics must face is to specify what role a truth theory is supposed to play within a meaning theory. The most detailed proposal for tackling this problem is the account developed by Ernest Lepore and Kirk Ludwig. However, as I will show in this paper, theories along the lines of Lepore and Ludwig do not suffice to put someone into the position to understand the objectlanguage. The fundamental problem of truth conditional semantics thus remains unsolved.     

Predication and cognitive context: Between minimalism and contextualism

In this paper, we suggest a strategy for modelling cognitive context within a truth‐conditional semantics, using Asher's model of predication. This allows us to introduce the notion of type presupposition intended as a lexical constraint to the composition of the truth‐conditional content. More specifically, we suggest that this model of predication produces a notion of truth‐conditional meaning where the cognitive context fixes a set of lexical restrictions and forced modifications. We conclude that this model might offer an intermediate position between Minimalism and Contextualism: an account that provides intuitive truth conditions within a formal semantic theory.     

Content in a Dynamic Context

The standing tradition in theorizing about meaning, since at least Frege (1882), identifies meaning with propositions, which are, or determine, the truth‐conditions of a sentence in a context. But a recent trend has advocated a departure from this tradition: in particular, it has been argued that modal claims do not express standard propositional contents. This non‐propositionalism has received different implementations in expressivist semantics (Moss, 2015; Swanson, 2006; Yalcin, 2007) and certain kinds of dynamic semantics (Gillies, 2004, 2010; von Fintel and Gillies, 2007; Veltman, 1985, 1996). They maintain that the key aspect of interpretation of modal claims is the characteristic dynamic effect they have on the context. I argue that pessimism about truth‐conditions arises from an overly simplistic picture of content, context and their interaction. While I agree with the critics that an important aspect of modal meaning is the dynamic effect modals have on the context, I argue that they have mischaracterized the nature and the complexity of this effect. A more nuanced account of the interaction between modals and context shows that far from being incompatible with propositional meaning, the dynamic aspect of meaning is precisely what allows us to predict the correct propositional content of an utterance.     

意义理论

Research into logical syntax provides us the knowledge of the structure of sentences, while logical semantics provides a window into uncovering the truth of sentences. Therefore, it is natural to make sentences and truth the central concern when one deals with the theory of meaning logically. Although their theories of meaning differ greatly, both Michael Dummett’s theory and Donald Davidson’s theory are concerned with sentences and truth and developed in terms of truth. Logical theories and methods first introduced by G. Frege underwent great developments during the past century and have played an important role in expanding these two scholars’ theories of meaning. Translated by Ma Minghui from Zhexue Yanjiu 哲学研究 (Philosophical Research), 2006, (7): 53–61     

Comparing Fixed-Point and Revision Theories of Truth

In response to the liar’s paradox, Kripke developed the fixed-point semantics for languages expressing their own truth concepts. (Martin and Woodruff independently developed this semantics, but not to the same extent as Kripke.) Kripke’s work suggests a number of related fixed-point theories of truth for such languages. Gupta and Belnap develop their revision theory of truth in contrast to the fixed-point theories. The current paper considers three natural ways to compare the various resulting theories of truth, and establishes the resulting relationships among these theories. The point is to get a sense of the lay of the land amid a variety of options. Our results will also provide technical fodder for the methodological remarks of the companion paper to this one.     

The dynamic lexicon in a truth-conditional framework; or how to have Your cake and eat it

A fundamental principle of all truth-conditional approaches to semantics is that the meanings of sentences of natural language can be compositionally specified in terms of truth conditions, where the meanings of the sentences’ parts (words/lexical items) are specified in terms of the contribution they make to such conditions their host sentences possess. Thus, meanings of words fit the meanings of sentences at least to the extent that the stability of what a sentence might mean as specified in a theory is in accord with the stability of what a word might mean as similarly specified. In this paper, I shall be concerned with Ludlow’s (2014) idea that, in fact, there need be no such sympathy between words and sentences. He proposes that we can square what he calls a dynamic lexicon, where word meaning is not stable at all, with a traditional truth-conditional approach of the kind indicated, where sentence meaning is delivered via ‘absolute truth conditions’. I share Ludlow’s aspiration to accommodate dynamic features of word meaning with a truth conditional approach, but not his belief that the marriage is an easy deal. Thus, I shall present a problem for Ludlow’s position and show how resolving this problem leads to an alternative picture of how the meaning of a sentence may be truth-conditionally specified with all relevant dynamic features of the lexicon retained.     

The assertion-candidate and the meaning of mood

The meaning of a declarative sentence and that of an interrogative sentence differ in their aspect of mood. A semantics of mood has to account for the differences in meaning between these sentences, and it also has to explain that sentences in different moods may have a common core. The meaning of the declarative mood is to be explained not in terms of actual force (contra Dummett), but in terms of potential force. The meaning of the declarative sentence (including its mood) is called the assertion-candidate, which is explained by what one must know in order to be entitled to utter the declarative with assertive force. Both a cognitive notion (knowledge) and a pragmatic notion (assertive force) are thus part of the explanation of the assertion-candidate. Davidson’s criticism that such a theory is in need of an account of the distinction between standard and non-standard uses of the declarative is answered: without counter-indications an utterance of a declarative sentence is understood as having assertive force. The meaning of an interrogative sentence, the question-candidate, and that of the other sentence types can ultimately be explained in terms of their specific relations to the assertion-candidate. Martin-Löf’s constructive type theory is used to show the philosophical relevance of a semantics of mood. The constructivist notion of proposition needs to be embedded in a theory of the assertion-candidate, which fulfils the offices of being the meaning of the declarative sentence, the content of judgement and assertion and the bearer of epistemic truth.     

Events,Tropes, and Truthmaking

Nominalizations are expressions that are particularly challenging philosophically in that they help form singular terms that seem to refer to abstract or derived objects often considered controversial. The three standard views about the semantics of nominalizations are [1] that they map mere meanings onto objects, [2] that they refer to implicit arguments, and [3] that they introduce new objects, in virtue of their compositional semantics. In the second case, nominalizations do not add anything new but pick up objects that would be present anyway in the semantic structure of a corresponding sentence without a nominalization. In the first and third case, nominalizations in a sense ‘create’ new objects’, enriching the ontology on the basis of the meaning of expressions. I will argue that there is a fourth kind of nominalization which requires a quite different treatment. These are nominalizations that introduce ‘new’ objects, but only partially characterize them. Such nominalizations generally refer to events or tropes. I will explore an account according on which such nominalizations refer to truth makers.     

On Value‐Attributions: Semantics and Beyond

This paper is driven by the idea that the contextualism‐relativism debate regarding the semantics of value‐attributions turns on certain extra‐semantic assumptions that are unwarranted. One is the assumption that the many‐place predicate of truth, deployed by compositional semantics, cannot be directly appealed to in theorizing about people's assessments of truth value but must be supplemented (if not replaced) by a different truth‐predicate, obtained through certain “postsemantic” principles. Another is the assumption that semantics assigns to sentences not only truth values (as a function of various parameters, such as contexts, worlds and times), but also semantic contents, and that what context‐sensitive expressions contribute to content are contextually determined elements. My first aim in this paper will be to show how the two assumptions have shaped two ways of understanding the debate between contextualism and relativism. My second aim will be to show that both assumptions belong outside semantics and are, moreover, questionable.     

Constructive Logic, Truth and Warranted Assertability

Shapiro and Taschek have argued that simply using intuitionistic logic and its Heyting semantics, one can show that there are no gaps in warranted assertability. That is, given that a discourse is faithfully modelled using Heyting's semantics for the logical constants, then if a statement S is not warrantedly assertable, its negation ^∼ S is. Tennant has argued for this conclusion on similar grounds. I show that these arguments fail, albeit in illuminating ways. An appeal to constructive logic does not commit one to this strong epistemological thesis, but appeals to semantics of intuitionistic logic none the less do give us certain conclusions about the connections between warranted assertability and truth.     

Color,context, and compositionality

Color adjectives have played a central role in work on language typology and variation, but there has been relatively little investigation of their meanings by researchers in formal semantics. This is surprising given the fact that color terms have been at the center of debates in the philosophy of language over foundational questions, in particular whether the idea of a compositional, truth-conditional theory of natural language semantics is even coherent. The challenge presented by color terms is articulated in detail in the work of Charles Travis. Travis argues that structurally isomorphic sentences containing color adjectives can shift truth value from context to context depending on how they are used and in the absence of effects of vagueness or ambiguity/polysemy, and concludes that a deterministic mapping from structures to truth conditions is impossible. The goal of this paper is to provide a linguistic perspective on this issue, which we believe defuses Travis’ challenge. We provide empirical arguments that color adjectives are in fact ambiguous between gradable and nongradable interpretations, and that this simple ambiguity, together with independently motivated options concerning scalar dimension within the gradable reading accounts for the Travis facts in a simpler, more constrained, and thus ultimately more successful fashion than recent contextualist analyses such as those in Szabó (Perspectives on semantics, pragmatics and discourse: A festschrift for Ferenc Kiefer, 2001) or Rothschild and Segal (Mind Lang, 2009).     

Play, agreement and consensus

In this paper I employ the analysis of play to clarify the distinction between an agreement and a consensus, and I argue that it is the conditions supplied by the playful process that enable us to partake in the recognition and creation of truth. Gadamer's hermeneutical truth, unlike propositional truth, speaks of the interpretive act whereby meaning is recognized. This interpretive recognition of meaning is described by Gadamer as an occurrence of interpretive play orgenuine understanding. Regarding Gadamer's conception of truth as modelled on agreement rather than consensus, I demonstrate that it is in the process of play that we achieve the making and maintaining of the conditions that are necessary for agreement, and that the nature of understanding is most clearly revealed by viewing it as a playful process.     

Truth,Meaning, and Circularity

It is often argued that the combination of deflationism about truth and the truth-conditional theory of meaning is impossible for reasons of circularity. I distinguish, and reject, two strains of circularity argument. Arguments of the first strain hold that the combination has a circular account of the order in which one comes to know the meaning of a sentence and comes to know its truth condition. I show that these arguments fail to identify any circularity. Arguments of the second strain hold that the combination has a circular explanation of the ideas or concepts of meaning and truth. I show that these arguments identify a genuine, but acceptable, circularity.     

Tolerant, Classical, Strict

In this paper we investigate a semantics for first-order logic originally proposed by R. van Rooij to account for the idea that vague predicates are tolerant, that is, for the principle that if x is P, then y should be P whenever y is similar enough to x. The semantics, which makes use of indifference relations to model similarity, rests on the interaction of three notions of truth: the classical notion, and two dual notions simultaneously defined in terms of it, which we call tolerant truth and strict truth. We characterize the space of consequence relations definable in terms of those and discuss the kind of solution this gives to the sorites paradox. We discuss some applications of the framework to the pragmatics and psycholinguistics of vague predicates, in particular regarding judgments about borderline cases.     

Supervaluations and the Propositional Attitude Constraint

For the sentences of languages that contain operators that express the concepts of definiteness and indefiniteness, there is an unavoidable tension between a truth-theoretic semantics that delivers truth conditions for those sentences that capture their propositional contents and any model-theoretic semantics that has a story to tell about how indetifiniteness in a constituent affects the semantic value of sentences which imbed it. But semantic theories of both kinds play essential roles, so the tension needs to be resolved. I argue that it is the truth theory which correctly characterises the notion of truth, per se. When we take into account the considerations required to bring model theory into harmony with truth theory, those considerations undermine the arguments standardly used to motivate supervaluational model theories designed to validate classical logic. But those considerations also show that celebration would be premature for advocates of the most frequently encountered rival approach – many-valued model theory.     

Play,agreement and consensus

In this paper I employ the analysis of play to clarify the distinction between an agreement and a consensus, and I argue that it is the conditions supplied by the playful process that enable us to partake in the recognition and creation of truth. Gadamer's hermeneutical truth, unlike propositional truth, speaks of the interpretive act whereby meaning is recognized. This interpretive recognition of meaning is described by Gadamer as an occurrence of interpretive play orgenuine understanding. Regarding Gadamer's conception of truth as modelled on agreement rather than consensus, I demonstrate that it is in the process of play that we achieve the making and maintaining of the conditions that are necessary for agreement, and that the nature of understanding is most clearly revealed by viewing it as a playful process.