09级7班 李梦琦 2009237
Aspects of Meaning Construction in Natural Language
Mental spaces is written by Gilles Fauconnier and published by Cambridge University in 1994.
This book is a major advance in the study of reference, descriptions, and coreference, which have long been at the center of research in linguistics and the philosophy of language. The more traditional theories assume that natural language semantics can be adequately studied with the tools of formal logic. Fauconnier, however, recognized that the tools of formal logic fail when confronted with the full range of natural language phenomena. He has realized that what is needed instead is a cognitive theory, a theory that is based on the capacities of the human mind rather than the capacities of the mathematical systems that happen to be used by logician.
Fauconnier posits a theory in which reference has a dimension of structure all its own, which is simply representable using mental spaces, connectors across the spaces, and a few general principles. The complexity lies largely in the interaction between the principles, and in the contextual structures that feed into the principles for interpretation. With such a simple structure, Fauconnier can handle examples that are beyond the capacities of complex logical theories. There is a salient class of such example, such as the split self, split coreference, the theory of logical form, possible worlds and situations, discourse representation theory, etc.
First is the methodology and empirical base. The methodology is a classic
scientific one: aim for the most encompassing generalizations; bring together theoretically superficially diverse empirical data; try not to prejudge theoretical outcomes with premature descriptive classifications. Second is the subject matter about meaning, method and form. A natural language sentence is cognitively complex, because it incorporates information and building instructions at all these different levels. What kind of meaning will be actually produced depends on the mental space configuration (generated by early discourse) that the sentence actually applies to. Third is the substance and formalization. Fourth is the development of mental space research. The theory of mental spaces and its application to problems of reference and presuppositions, some classical, some novel, was first presented in 1978 at the Academia della Crusca, in Florence. In later years, the approach was applied and extended to many areas that had not been foreseen, suggesting that mental spaces and mental space connections were fat more pervasive than we had first imagined.
This book is divided into five chapters. Chapter 1 is about pragmatic functions and images. In the quest for a fully explicit and maximally integrated account of language organization, much attention has been focused on the multi-level structural intricacies of linguistic forms. However, some studies have shifted this focus of attention from the language forms themselves to other structures and networks on which they depend and to the correspondences that hold, or are established, between such structures and networks. Outstanding examples are the notions of frames and scenarios; literal metaphor as an elaborate structuring of conceptual networks via partial correspondences underlying semantic-pragmatic organization and its
expression through language; the account of presupposition in terms of discourse worlds linked to each other; and the treatment of “scopal” phenomena like opacity
and transparency as referential correspondence between concrete or mental images. Chapter 2 talks about roles and multiple connectors, which suggest that, among other things, mental spaces provide a linguistic and cognitive means for setting up and distinguishing frames. Chapter 3 is about presuppositions: floating, transfer and projection strategies. Chapter 4 discusses counterfactuals and comparatives. Traditionally, counterfactuals are viewed as cases of possibly valid reasoning from premises that are false in actuality. Attempts to evaluate the truth conditions of counterfactuals involve, among others, two general kinds of questions, the problem of determining which true statements are combined with the false premises to carry out the reasoning and the problem of determining when and which logical laws apply to counterfactuals. Chapter 5 is about transspatial operators, philosophical issues and future perspectives. Speech q, discourse reference, quantification, and generics are traditional problem areas for which the mental space perspective suggests fresh research orientations. One could also investigate reported speech, the multiple special layers used to construct narratives, the social schemata that structures spaces and set up certain pragmatic connectors rather than others, or the effects of a gestural modality, instead of an oral one, on the implementation of space phenomena. Sign language seems to have ways of setting up spaces and elements (i.e. abstract referential domains) using body-shift and three-dimensionality. The mental construction, which remains the same regardless of the modality involved, can be
reflected concretely by very different codes, adapted to that modality.
A possible world, in formal logic, is a state description: a set of entities and the properties and relations that hold of them. Possible world semantics is a formally stated relationship between expression in some formal language and set-theoretical models of possible worlds or situations. These are objectivist models, models of the actual world, or a possible world, or an actual or possible situation. Possible worlds and situations are not models of the human mind, but models of the world as it is assumed to be or might be. Discourse representation theory is developed after Fauconnier, which is similar to situation semantics in that it includes partial state descriptions. It adds one important thing: the ability to augment a state description as one goes through a discourse. The other accomplishment of Fauconnier’s theory is
that it demonstrates that the problem of presupposition inheritance is the same as the problem of reference, and that the same theory works for both.
This book goes through the most difficult known problems in reference and presupposition, and solves them all with equal ease. It also meshes with all the research done on conceptual systems within the field of cognitive semantics. Fauconnier’s theory is truly elegant. It has become the standard account for these problems within cognitive linguistics. This reissue of Fauconnier’s already classic
work is most welcome.