The Unity of Science is a compact philosophical essay in which Rudolf Carnap sets out the logical-empiricist program for a unified science. Its central contention is that the sciences are not separate realms with inaccessible subject matters, but form a single system of statements because the empirical statements of all the sciences can be expressed in a common language—the language of physics. The unity at issue is primarily a unity of language and logical method, not a metaphysical claim that everything is “really” physical. Carnap’s physicalism is linguistic: the physical language is to serve as the universal language of science because it is intersubjective, spatiotemporal, and suited to testable observation.
The book begins from the distinction between scientific knowledge and metaphysical pseudo-statement. Philosophy, on this account, is not a special science with its own objects of knowledge; it is the logical analysis of the language of science. Its task is to clarify the logical form of scientific statements, remove confusions, and show how those statements are connected to observation and test. Statements that cannot be brought into relation with possible observation, or reconstructed through logical analysis, are rejected as pseudo-problems. This polemic against metaphysics is not an incidental feature of the work but part of its conception of what philosophy can legitimately do.
Carnap develops the thesis that the physical language is universal. Statements in biology, psychology, and other empirical disciplines can be translated into, or reduced to, statements formulated in physical terms. The point is not that every science is already stated in the vocabulary of physics, but that its terms and laws can be logically connected with physical predicates and observation statements. This makes possible a unified system of science rather than a collection of discontinuous languages. Carnap’s treatment of psychology is especially important: statements about mental states are not private reports of an inaccessible inner world. They can be reformulated in terms of observable behavior and dispositions to behave, and thus brought into the public, intersubjective language of science. In this way the unity of science also answers the worry that psychological or “human” sciences must remain methodologically separate.
A major problem the book addresses is how observation statements can serve verification without becoming private to the individual observer. If scientific observation were merely a matter of private sensation, science would lack an intersubjective basis. Carnap therefore argues that the observation statements required for verification—protocol sentences—are not private to the observer. They are formulated in public, physical language and can be checked, repeated, and criticized by others. They are not absolute and incorrigible foundations; they are part of the scientific system and can be revised in the light of further evidence. The unity of science depends on this intersubjective testability, and the physical language provides the medium in which such testability is possible.
The work also presents the unity of science as a program of logical reconstruction rather than as a description of how the sciences happen already to be organized. Higher-level scientific concepts are to be connected by chains of reduction to lower-level, ultimately physical, concepts. The result would be a hierarchy of concepts and laws forming one system. Carnap does not claim that this reconstruction is complete; the unity of science is a regulative ideal and a research program. It directs attention to the logical connections among the sciences and to the possibility of translating their statements into a common language.
Throughout, the book opposes the idea that different sciences have fundamentally different logical methods or inaccessible essences. Differences among the sciences are differences of subject matter, degree of abstraction, and stage of development, not differences of logical type. All empirical science employs observation, logical systematization, and intersubjective testing. The physical language can accommodate all empirical statements, and philosophy’s role is to clarify that language rather than to add metaphysical doctrines about reality. The Unity of Science is thus a short but programmatic statement of logical empiricism: science is unified through logical analysis and a universal physical language; philosophy is the analysis of scientific language; metaphysics is rejected as pseudo-problem; observation statements are public and revisable; and the several sciences are to be understood as parts of one system of knowledge.
An important qualification in the book is that its “physicalism” is linguistic and methodological rather than ontological. The claim is not that only physical entities exist or that psychological and biological discourse is illusory, but that the language of observable spatiotemporal things and events can serve as the universal medium in which empirical statements are made intersubjectively checkable. Physical language is thus a condition of unified science, not a metaphysical inventory of what there is.
Likewise, the unity of science is not an eliminationist demand that all disciplines be replaced by physics. The reduction of higher-level concepts to physical ones is meant to preserve the distinctive subject matters and laws of the special sciences while displaying their logical connections to a common evidential base. The programme is integrative and critical at once: it asks of any scientific claim how it is tested and how it fits into the system of knowledge, and it treats divisions among the sciences as problems of translation, reduction, and method rather than as ultimate divisions of reality.
