1. Grice theory: the development from "classical" to "new"
Grice's conversational implicature theory has made great contributions to linguistics and logic, that is, it has developed a new form of reasoning: pragmatic reasoning. At present, the study of conversational implicature has developed from "classic Grice's theory of conversational implicature" to "new Grice's theory of conversational implicature", which is a new impetus to the study of pragmatic inference.
The academic value, practical significance and some shortcomings of the classical Grignard theory are not repeated here, except for one thing: when Grignard constructed the "cooperative principle", he did not have time to propose a mechanism to deduce the conversational implicature by using this principle. Therefore, I wrote Pragmatic Inference (Journal of Foreign Languages 199 1 6), trying to give this mechanism and related laws.
Gerhardt made a special speech in the 1960s and published an article in the 1970s. Since then, researchers have sprung up like mushrooms after rain, which has greatly developed the theory of conversational implicature. Scholars have made two efforts to make up for the deficiency of the cooperative principle: (1) applying the politeness principle, the salvation principle and the cooperative principle; (2) Reforming and reconstructing the new principle of conversational implicature. For the first aspect, China scholars have summarized many articles. For the second aspect, the response of domestic scholars is still rare.
There are two main ways to reform and reconstruct the principle of conversational implicature: one is to replace the cooperative principle with a single "relevance principle", which was put forward by Sperber and Wilson (1986). The second is to concretize standards, especially quantitative standards. Many scholars have made great achievements, such as Gatesta, Atlas, Horn and Joseph Richmond Levenson. Atlas and Joseph Richmond Levenson have put forward a relatively complete concept in Atlas and levinson 198 1. Later, Joseph Richmond Levenson expounded in "Levinson, 1983" that the quantitative standard can be reconstructed from two aspects: based on the 1 th order of the quantitative standard, the hierarchical meaning and the clause meaning are deduced; Under the guidance of the second rule, we can invite more information from less information, which can be called "information principle". This is actually the actual content of the first two principles in Joseph Richmond Levenson's Three Principles. 1987, Joseph Richmond Levenson put forward his three principles of conversational implicature; Liege himself said that this is a neoclassical interpretation of Grignard's criterion. It is generally believed that the development of Grignard's theory from "classical" to "new" has now moved from the gestation period to the transition period. 199 1 year, Joseph Richmond Levenson officially called his three principles "New Grice Pragmatic Instruments", and he also used "classicism" to explain his theory. (See Levinson, 1987, 199, respectively), some scholars began to use the terms "classical-"and "new-"in this sense.
Second, Neo-Grice's Conversational Implicature Theory: Three Principles of Liesl
The main points of Joseph Richmond Levenson's "New Grice Pragmatic Mechanism" are as follows:
Quantity principle
1, speaker standard:
Don't let your statement be weaker in information than your knowledge allows, unless a stronger statement conflicts with the information principle.
Recipient's inference:
Believe that the speaker has provided the most powerful information he knows, so:
(1) When the speaker says A(W), it forms a "Horn's hierarchy" (hereinafter referred to as "Horn's relation", so the author presses it), even A(S)├A(W), then k ~ (a (s)) can be deduced, that is to say, the speaker knows.
(2) The speaker speaks A(W), but A(W) does not contain the content of embedded sentence Q, but the content of Q is contained in A(S), which has strong information, and SW forms a comparison set, so ~ k (q) can be deduced, that is, the speaker does not know whether Q can be established. (Author Press: The following explains the meaning of these symbols and expressions one after another. )
2. Information principle
Speaker standard: minimum standard
"Speak as little as possible", that is, only provide the least language information, as long as the communicative purpose can be achieved;
Audience inference: extended rules
Expand the information content of the speaker's discourse by finding out the specific understanding until it is recognized as the semantic intention of the speaker. In particular:
(1) The relationship between the object and the event mentioned in the setting sentence is conventional, unless a. This is inconsistent with the confirmed situation; B. The speaker violated the minimum standards and used lengthy expressions;
(2) If some existence or fact is consistent with the confirmed situation, it is assumed that this is exactly what the sentence is about to say. 3. Mode principle
Speaker standard:
Don't use long, obscure or marked expressions for no reason.
Recipient's inference:
The speaker uses a long marked expression, and his meaning is different from what he could have expressed in unmarked form, especially when he tries to avoid conventional association or deduces the meaning of unmarked expression by information principle.
Regarding the principle of quantity, I want to talk about the "Holstein relationship" here. Horn believes that there are differences in semantic information strength between words in groups like the following: