Carnap's Philosophical Characteristics

Carnap's philosophy starts from the following basic beliefs:

(1) In science and scientific methods, it is considered that science is the only systematic and well-founded knowledge possessed by human beings, and scientific methods are the only way for people to acquire this knowledge.

② In the aspect of mathematical logic, it is considered that the logical analysis method initiated by Frege and Russell has brought philosophy into a new stage and is the only effective method to study philosophy.

The ultimate goal of philosophy is to eliminate the confusion and mistakes of the old philosophy by using logical analysis methods, and to logically clarify the empirical basis of science.

Carnap's philosophy has the following main features:

All concepts and knowledge about the world ultimately come from direct experience.

(2) Philosophical problems come down to language problems, philosophy is the logic of scientific language, and the philosophical method lies in the logical analysis of scientific language.

③ Inductive reasoning can and should be as standardized and accurate as deductive reasoning, and inductive logic provides a standard for evaluating people's reasonable beliefs and behaviors.

The foundation of empirical knowledge

Carnap agrees with the dichotomy of analytical proposition and comprehensive proposition, which has been generally accepted by empiricism since david hume. He believes that mathematics and logic propositions are analytical propositions, and comprehensive propositions come from experience. The most basic empirical proposition is the original recorded statement or the original recorded statement, that is, the statement that records the direct observation results. He pointed out that to clarify the basis of empirical knowledge, it is not necessary to examine the actual process of human acquisition of knowledge, but to explore how to restore scientific propositions and concepts to those that record direct experience. In this way, classical epistemology, which studies cognitive processes from a psychological perspective, is ruled out.

Carnap tried to establish a strict logical system to represent the domain of all knowable objects. This logical combination system can simplify a field into a few concepts through a series of combination definitions. Carnap divided the known objects into four categories: ① social and cultural objects; 2 other people's hearts; 3 physical objects; (4) my own experience. He chose ④ as the basis of epistemology, among which he chose the simplest and most primitive symmetrical relation, namely Er relation, which refers to the memory of similarity. Basic experience can be defined as the field of er relations. He used partial similarity to express the relationship between the same experiences, that is, two-way Er relationship; Similar regions are abstract classes with partial similarity. In this way, by applying more complex definitions, we can further construct the concepts of feeling and time sequence, and then define the world of time and space, the world of perception, the biological world, human beings and human cultural objects.

For Carnap. The choice of solipsism foundation has only methodological significance, but no ontological significance. This can only show that Carnap's early epistemology tends to phenomenology advocated by Mach and Russell. He believes that logically speaking, you can also choose physical objects as the basis. Carnap pointed out that different choices only mean different languages, and the original recorded sentences are different in nature and form. He is neutral to the debate of different philosophical schools, and thinks that people can freely choose their favorite language. The difference between idealism and realism, phenomenology and physicalism is only the result of philosophers choosing different languages.

Later, under the influence of Newrat, Carnap turned to physicalism, thinking that physical language is the most desirable initial sentence form, because physical language can not only express all knowledge (including psychological knowledge), but also have mutual observation. Therefore, he accepted and actively participated in the work of unifying science proposed by Newrat, trying to unify social science, humanities and natural science on the basis of physical language.

Simplification of theoretical concepts

Carnap believes that theoretical concepts can be reduced to a set of observation concepts through clear definitions, and the reduced concepts are equivalent to this set of reduced concepts. This reduction means that theoretical concepts can be replaced by observation concepts, thus becoming a strict empiricism. Later, he realized that reduction such as solubility and brittleness could not be realized and could not be clearly defined by direct experience, so he relaxed this requirement and stipulated another reduction sentence. This kind of reduction sentence takes the form of conditional sentence, which gives the connection between the concept of reduction and direct experience in a specific situation. The difference between reduction statement and definition reduction is that the former can not completely reduce the theoretical statement to the initial statement representing direct experience, but can only deduce various observation statements that change according to different experimental conditions. This fundamentally shakes the principle of early logical positivism as a meaningful proposition standard. Carnap replaced it with the principle of confirmation and the principle of inspection, and thought that although the scientific hypothesis about the unobservable object can never be confirmed by the observational evidence, it can be confirmed by the observational evidence to some extent. The quantitative study of the degree of confirmation prompted Carnap to engage in the work of logical probability in his later period. In the basic research of natural science, Carnap put forward a two-layer language theory about theoretical structure. He divides scientific language into observation language and theoretical language, and thinks that theoretical language can only be indirectly and partially explained by corresponding rules, and the meaning of theoretical concepts depends on the theory itself. In this way, Carnap's definition of positivism has been further relaxed, indicating that he has moved away from his original positivism.

artificial language

Carnap distinguished object language and metalanguage in language analysis. Meta-language can express the theory of object language, which is meta-logic, and logical grammar is a part of it. Carnap tried to make metalanguage precise and construct an accurate conceptual system of logical grammar. Carnap strictly limited logical grammar to the form of language expression, not to its meaning. In his view, logical grammar includes formation rules and deformation rules. The former is used to specify the symbols used and how the symbols form sentences, while the latter refers to the rules of logical reasoning. Carnap thinks that logical rules only involve the form of language, and given these two rules, the language is given. The artificial language constructed by strictly following these rules can accurately realize the logical reconstruction of knowledge. Carnap put forward the principle of tolerance in language choice, that is, people can freely choose different rule systems, thus choosing different logics and languages.

Carnap distinguished substantive ways of speaking from formal ways of speaking. He pointed out that in natural language, some sentences about grammar are very similar to those about objects in form, so they are mistaken for object sentences. Carnap regards these statements as pseudo-object statements, and thinks that they are an important source of pseudo-problems in philosophy. In his view, this kind of mistake is due to people's wrong use of substantive ways to express what should have been expressed by formal ways. Once they are converted into formal methods, that is, grammatical sentences translated into metalanguage, confusion and mistakes can be eliminated.

In his early days, Carnap thought that philosophical problems were only grammatical problems, and meta-theory only included grammar. Later, influenced by A. Talsky, I realized that philosophy should also include semantic analysis, and regarded semantics as a part of meta-theory. Carnap's basic concepts about semantics are factual truth and logical truth. He also distinguishes connotation from extension, and thinks that the basic concepts of connotation semantics are the meaning of names and sentences, the synonymy of expressions and the analysis of propositions; The basic concepts of extended semantics processing are individual, class, truth value and so on. On this basis, Carnap explained the concept of modal logic with the concept of semantics, and interpreted the modal of a proposition as the semantic nature of the corresponding sentence expressing the proposition.

Logical probability and inductive logic

In order to revise his earlier confirmation principle, Carnap replaced it with confirmation principle, and thought that the relationship between scientific knowledge and direct experience was that scientific hypothesis was confirmed to some extent from empirical evidence. The central concept of this theory is the degree of confirmation, that is, logical probability or inductive probability. The so-called logical probability is to regard probability as a logical concept, which is different from the statistical probability based on relative frequency. Carnap believes that logical probability is the basis of all inductive reasoning without deductive inevitability, and the theory about logical probability is inductive logic. He pointed out that logical probability does not belong to science itself, but belongs to scientific logic, that is, scientific methodology, which should be expressed in metalanguage. The characteristic of logical probability is that its probability value depends on the logical relationship between sentences and has nothing to do with experience. The statement of logical probability is an analytical statement, which determines the quantitative confirmation relationship between statements, similar to deductive logic, so inductive logic can give the confirmation degree of hypothesis relative to given evidence. Therefore, he thinks there is no fundamental difference between inductive logic and deductive logic, inductive reasoning and deductive reasoning.

Carnap conceived a set of formal inductive logic, including defined confirmation function and general estimation theory. He also explained the concept of information semantically by using the idea of inductive logic and combining the semantic concept of sentence content.

There are many technical difficulties in Carnap's research on inductive logic, and his own work in this field is far from reaching his own goal. There are many people in western philosophy who are skeptical about his ideas.