21 of 22 people found the following review helpful:
Difficult reading but gives a different perspective., 2004-04-25
Cognitive grammar is presented in this book as a theory that is "fundamentally at odds with the dominant trends in current linguistic theory", especially the generative tradition of grammar. The author's main thesis is the inseparability of syntax and semantics, and he attempts to present an account of linguistic structure that is independent of the any particular domain. The formalism of accepted linguistic theory is to be rejected, in favor of what the author calls a "conceptual clarification" of the fundamental issues in linguistics. Figurative language, with its idioms, metaphors, and "semantic extensions" is to be given central status, the author argues. For this reviewer, the book was difficult reading, this no doubt due to its departure from the standard formal constructions of linguistics. The author recognizes the difficulty he faces in presenting his theory, he cautions the reader that the theory is not complete, but he asserts that the basic elements of the theory are "stable" enough to allow presentation in a work such as this. These basic elements are summarized as saying that semantic structure is not universal, but specific to a certain language; grammar is symbolic in nature and consists of the symbolization of semantic structure; and that grammar and lexicon cannot be distinguished.
Readers with a background in cognitive science or neuroscience may be taken aback by the author's carefree talk of the functioning of the human brain and mental events, which is frequently done with no empirical evidence to substantiate it. The author though is aware of the lack of evidence, for he says early on that some of the book could be regarded as an exercise in what he calls "speculative psychology." He recognizes that his discussion of the cognitive workings of the human mind may not be accurate, but he believes that they are plausible and that the overall system he presents has enough coherence to be valid in many respects.
The "general assumptions" of cognitive grammar are that of "symbolization", namely that language is symbolic in nature and consists of an open-ended set of linguistic expressions. These expressions associate a semantic representation of some kind with a phonological representation, but this association is not entirely arbitrary. Grammar is the process of grouping morphemes into progressively larger configurations and is inherently symbolic. Grammatical and semantic components are not to be separated, for grammar is the structuring and symbolization of semantic content. The second assumption is that language is an essential part of human cognition, and that if one is to understand linguistic structure one needs to understand cognitive processing. There is no well-defined separation between linguistic ability and other forms of cognitive processing, and therefore cognitive psychology and linguistics must be integrated. The third assumption is that of `naturalness': descriptions must be `natural' in that they must deal with data in all its complexity and subtlety. Natural descriptions are to be contrasted with `artifactual' ones, which destroy the intrinsic organization of the data. Linguistic data is rarely discrete and linguistic categories are not always well defined. A linguistic theory must be `substantive', i.e. such a theory must possess conceptual clarity, and it must also correspond to something real.
For the author a `grammar' is a comprehensive description of its structure, with one class of grammars being those that can be written, and another being those that are envisioned by linguists. The latter do not exist he says, since they must satisfy requirements that cannot be met, such as being exhaustive in their coverage and psychologically accurate. The author though views cognitive grammar as a body of knowledge that can be constructed, that being done in terms of what he calls a `structured inventory of conventional linguistic units.' These units are the structures that have been mastered by the speaker to such an extent that they are habitual, and even though may have a large degree of `internal complexity', they are `effectively simple' in that they do not require constructive effort for their use.
Linguistic structures are organized on a continuous scale of what the author calls `entrenchment'. The regular use of a structure moves up its level of entrenchment, but long periods of disuse move it down. A novel structure can become so entrenched that it finally becomes a unit. Each unit has different degrees of entrenchment depending on its frequency of occurrence. The author asserts therefore that the greater the entrenchment the greater the linguistic significance.
The system of cognitive grammar is summarized as saying that semantic units are defined relative to cognitive `domains'. A concept or a knowledge base can function as one of these domains. In addition, linguistic semantics is asserted to be `encyclopedic', in that the meaning of an expression is usually dependent on specifications in many domains. Some of these domains play a more central role than others in giving this meaning. However, the role of expression in various domains can frequently be a peripheral aspect of its semantic value. The author encapsulates this by saying that language can `self-reference'. Lastly, the units are acquired through `decontextualization'. This means that through a process of abstraction, the extent of which is determined by the varieties of settings of the units and how consistently the units appear in these settings. The grammar of a language is to be viewed as an `inventory of linguistic units' and not a "generative description" that proceeds by formally constructing the well-formed sentences of the language. It is also not a set of operations what output the well-formed sentences. Language is not an autonomous formal system, but rather can only be understood and characterized in the context of a theory of human cognition. A complete description of human cognition would be needed to give an exhaustive description of language.
19 of 24 people found the following review helpful:
Langacker's Cognitive Approach to Linguistics, 2000-12-10
Ronald Langacker Cognitive Grammar is one of the most fine works on language and cognition of the last twenty years. Instead of deducing the organisation of conceptual system by means of pragmatic analisys of language (as Jackendoff and Talmy for istance)Langacker take up the hard task of describing language as a natural-cognitive based phenomena. Language, says Langacker, is simbolic in nature. This means that meaning pertains to all levels of language structure and that sintax and semantics can not be studied as separate components of language as argued by Chomsky. Moreover, semantics is strictly endowned in cognive structure. The cognive system and the cognive abilities of mind are, therefore, basical to the understanding of conceptual and linguistic structures. Again, the central role played by the notions of Landmark and Trajectory in Langacker'Cognitive Grammar lead Langacker's theory very close to gestalt psychology and to the struttural approach to language such that of Tesniére [1959]. All those aspects make of Langacker's CG a masterpiece to be remembered in the History of Linguistics.