Predication in Conceptual Realism

Conceptual realism begins with a conceptualist theory of the nexus of predication in our speech and mental acts, a theory that explains the unity of those acts in terms of their referential and predicable aspects. This theory also contains as an integral part an intensional realism based on predicate nominalization and a reflexive abstraction in which the intensional contents of our concepts are “object”-ified, and by which an analysis of predication with intensional verbs can be given. Through a second nominalization of the common names that are part of conceptual realism’s theory of reference (via quantifier phrases), the theory also accounts for both plural reference and predication and mass noun reference and predication. Finally, a separate nexus of predication based on natural kinds and the natural properties and relations nomologically related to those natural kinds, is also an integral part of the framework of conceptual realism.     

Processing reflexives and pronouns in picture noun phrase

Binding theory (e.g., Chomsky, 1981) has played a central role in both syntactic theory and models of language processing. Its constraints are designed to predict that the referential domains of pronouns and reflexives are nonoverlapping, that is, are complementary; these constraints are also thought to play a role in online reference resolution. The predictions of binding theory and its role in sentence processing were tested in four experiments that monitored participants' eye movements as they followed spoken instructions to have a doll touch a picture belonging to another doll. The instructions used pronouns and reflexives embedded in picture noun phrases (PNPs) containing possessor phrases (e.g., Pick up Ken. Have Ken touch Harry's picture of himself). Although the interpretations assigned to pronouns were generally consistent with binding theory, reflexives were frequently assigned interpretations that violated binding theory. In addition, the timing and pattern of eye movements were inconsistent with models of language processing that assume that binding theory acts as an early filter to restrict the referential domain. The interpretations assigned to reflexives in PNPs with possessors suggest that they are binding-theory-exempt logophors, a conclusion that unifies the treatment of reflexives in PNPs.     

λ-Normal forms in an intensional logic for English

Montague [7] translates English into a tensed intensional logic, an extension of the typed -calculus. We prove that each translation reduces to a formula without -applications, unique to within change of bound variable. The proof has two main steps. We first prove that translations of English phrases have the special property that arguments to functions are modally closed. We then show that formulas in which arguments are modally closed have a unique fully reduced -normal form. As a corollary, translations of English phrases are contained in a simply defined proper subclass of the formulas of the intensional logic.This research is supported in part by National Science Foundation Grant BNS 76-23840.     

Saving the intuitions: polylithic reference

My main aim in this paper is to clarify the concepts of referential success and of referential continuity that are so crucial to the scientific realism debate. I start by considering the three dominant theories of reference and the intuitions that motivate each of them. Since several intuitions cited in support of one theory conflict with intuitions cited in support of another something has to give way. The traditional policy has been to reject all intuitions that clash with a chosen theory. A more radical policy, tied to some experimental philosophers, has called for the rejection of any evidential role for intuitions. I explore a largely ignored third alternative, i.e. saving intuitions (and their evidential role) even when they are at odds. To accommodate conflicting intuitions different sets of internally consistent (yet externally inconsistent) intuitions are taken to lend credence to different concepts of reference. In the current context, this means that the concepts of referential success and referential continuity are not monolithic. They are what I call ‘polylithic’. This paper is as much about meta-philosophical concerns with the role of intuitions as it is about reference and the scientific realism debate. Regarding the former I hope that a blueprint will emerge for similar projects in other philosophical domains. Regarding the latter, I hope that polylithicity helps disentangle claims about referential success and continuity in the scientific realism debate by making perspicuous which concepts are best equipped to evaluate the realist’s epistemic claims against the historical record of science.     

Comprehending Complex Concepts

Recent theories of concepts have raised the issue of how people combine simple concepts (like engine and repair) to form complex concepts (like engine repair). This article approaches this issue by asking how people comprehend modified noun phrases of this sort. One explanation of how complex concepts are understood (the feature weighting model) provides a simple mechanism in which the primary feature of the modifying concept is made more salient in the modified concept. Another explanation focuses on how world knowledge directs the combination process. The two explanations are compared in their ability to account for the interpretation of various kinds of noun phrases. Two experiments are reported which evaluate the feature weighting model's predictions for adjective-noun phrases. These contrasts suggest that the combination process does require reference to world knowledge. The consequences of accepting such an account are discussed.     

Processing doubly quantified sentences: Evidence from eye movements

We investigated the processing of doubly quantified sentences, such asKelly showed a photo to every critic, that are ambiguous as to whether the indefinite (a photo) specifies single or multiple referents. Ambiguity resolution requires the computation of relative quantifier scope: Whether a or every takes wide scope, thereby determining how many entities or events are to be represented. In an eye-tracking experiment, we manipulated quantifier order and whether continuations were singular or plural, for constructions with the direct or the indirect object occurring first. We obtained effects consistent with the on-line processing of relative scope at the doubly quantified phrase and considered two possible explanations for a preference for singular continuations to the quantified sentence. We conclude that relative quantifier scope is computed on line during reading but may not be a prerequisite for the resolution of definite anaphors, unless required by secondary tasks.     

Heterogeneous logic

This paper considers the question: what becomes of the notion of a logic as a way of codifying valid arguments when the customary assumption is dropped that the premisses and conclusions of these arguments are statements from some single language? An elegant treatment of the notion of a logic, when this assumption is in force, is that provided by Dana Scott's theory of consequence relations; this treatment is appropriately generalized in the present paper to the case where we do not make this assumption of linguistic homogeneity. Several applications of the resulting concept of a heterogeneous logic are suggested, but the main emphasis is on the formal development. One topic touched on is a certain contrast between the boolean and the intensional sentence-connectives in this more general setting.     

Many Many Problems

Supervaluationist solutions to 'the problem of the many' typically rely on two principles. First, the root of the problem is that singular terms can be vague, just as predicates can be. Secondly, the same resolution as the supervaluationist suggests for puzzles with vague predicates will resolve puzzles concerning vague singular terms. In recent years this second principle has been attacked from a number of fronts: it has been claimed that supervaluationist accounts of vague singular terms cannot explain the role of vague singular terms in propositional attitude reports, cannot explain penumbral connections between distinct singular terms, and cannot allow that vague singular terms are directly referential. While some of these arguments are successful against extant supervaluationist theories, I offer some natural modifications to supervaluationism which avoid the challenges.     

Leibniz on Intension,Extension, and the Representation of Syllogistic Inference

New light is shed on Leibniz’s commitment to the metaphysical priority of the intensional interpretation of logic by considering the arithmetical and graphical representations of syllogistic inference that Leibniz studied. Crucial to understanding this connection is the idea that concepts can be intensionally represented in terms of properties of geometric extension, though significantly not the simple geometric property of part-whole inclusion. I go on to provide an explanation for how Leibniz could maintain the metaphysical priority of the intensional interpretation while holding that logically the intensional and the extensional stand in strictly inverse relation to each other. This revised version was published online in June 2006 with corrections to the Cover Date.     

Natural language semantics in biproduct dagger categories

Biproduct dagger categories serve as models for natural language. In particular, the biproduct dagger category of finite dimensional vector spaces over the field of real numbers accommodates both the extensional models of predicate calculus and the intensional models of quantum logic. The morphisms representing the extensional meanings of a grammatical string are translated to morphisms representing the intensional meanings such that truth is preserved. Pregroup grammars serve as the tool that transforms a grammatical string into a morphism. The chosen linguistic examples concern negation, relative noun phrases, comprehension and quantifiers.     

Why do Alzheimer patients have difficulty with pronouns? Working memory, semantics, and reference in comprehension and production in Alzheimer's disease

Three experiments investigated the extent to which semantic and working-memory deficits contribute to Alzheimer patients' impairments in producing and comprehending referring expressions. In Experiment 1, the spontaneous speech of 11 patients with Alzheimer's disease (AD) contained a greater ratio of pronouns to full noun phrases than did the spontaneous speech produced by 9 healthy controls. Experiments 2 and 3 used a cross-modal naming methodology to compare reference comprehension in another group of 10 patients and 10 age-matched controls. In Experiment 2, patients were less sensitive than healthy controls to the grammatical information necessary for processing pronouns. In Experiment 3, patients were better able to remember referent information in short paragraphs when reference was maintained with full noun phrases rather than pronouns, but healthy controls showed the reverse pattern. Performance in all three experiments was linked to working memory performance but not to word finding difficulty. We discuss these findings in terms of a theory of reference processing, the Informational Load Hypothesis, which views referential impairments in AD as the consequence of normal discourse processing in the context of a working memory impairment.     

话语参照语境条件下汉语歧义短语的加工

采用移动窗口和眼动记录两种范式 ,以符合“VP +N1+的 +N2”格式的均衡型和述宾型两类汉语歧义短语为背景 ,在两个实验中 ,考察了话语参照语境影响歧义短语句法分析的机制以及时间进程。结果发现 ,在早于解歧区 1的区段上就开始出现话语语境效应。这些发现表明话语参照语境可以通过概念期望机制起作用 ,而并非仅仅通过参照前提机制起作用。此外 ,实验还证明了话语参照语境在句子加工早期的作用。这一结果为句法歧义消解的参照理论和基于制约的模型提供了证据 ,而对宣称最初的句法分析独立于话语语境的花园路径模型 ,提出了质疑     

Minimizing indexicality

I critically examine Cappelen and Lepore’s definition of and tests for indexicality, and refine them to improve their adequacy. Indexicals cannot be defined as expressions with different referents in different contexts unless linguistic meaning and circumstances of evaluation are held constant. I show that despite Cappelen and Lepore’s claim that there are only a handful of indexical expressions, their “basic set” includes a number of large and open classes, and generates an infinity of indexical phrases. And while the tests can be used effectively to combat contextualism concerning ‘knows’ and ‘actual,’ many expressions not in their basic set test positive for indexicality, including quantifier nouns, weather reports, and comparative adjectives. I rebut their claim that context-shifting arguments inevitably lead to radical contextualism, and that if there were any indexicals beyond their basic set, communication would be impossible.