The hypothesis (from the Greek ὑπόθεσις – the basis, the assumption) is a scientific assumption or assumption, the truth value of which is vague. Distinguish the hypothesis as a method for the development of scientific knowledge, which includes the advancement and subsequent experimental verification of assumptions, and as a structural element of the scientific theory.

The origin of the method of hypotheses is historically associated with the early stages of the development of ancient mathematics. Ancient Greek mathematicians widely used as a method of mathematical proof a deductive thought experiment, which included the promotion of hypotheses and the derivation from them through analytic deduction of the consequences to verify the correctness of the initial guesses. A fundamentally different approach to the hypothesis was proposed by Plato, who viewed it as a premise of the analytic-synthetic method of proof developed by him, capable of providing an absolutely true character of the derivation. Such an understanding of the heuristic role of the hypothesis was rejected by Aristotle, whose conception proceeded from the impossibility of using hypotheses as the premises of a syllogistic proof (since the only general, necessary and absolute truths were thought of as the latter), which led to a subsequent negative attitude to hypotheses as a form of unreliable or probable knowledge. In ancient science and natural science of the New Time, the method of hypotheses was used mainly only in an implicit, latent form within the framework of other methods (in thought experiments, in genetically constructive and inductive methods). This is evidenced by the “Elements” of Euclid and the statics of Archimedes, as well as the history of the formation of the mechanics of Galileo, Newton’s theory, molecular-kinetic theories. Only in the methodology and philosophy of the late 17th – early 19th centuries, in the process of comprehending the successes of empirical research, the heuristic role of the hypothesis method was gradually realized. However, neither the rationalistic nor the empirical directions in classical methodology and philosophy have been able to substantiate the necessity of hypotheses in scientific cognition and overcome the opposition of hypotheses and law. Thus, for example, Kant limited the sphere of application of scientific hypotheses to a narrow field of purely empirical research, attributing to the method of hypotheses an auxiliary, subordinate status in relation to a priori knowledge as knowledge of unconditionally universal and necessary truths.

In the seventies and eighties of the 19th century, F. Engels, on the basis of a fundamentally new understanding of the epistemological status of laws and theories as concerning the true assertions of a limited community, substantiated the role of scientific hypotheses not only in the process of accumulation and systematization of empirical material, but also at the stages of refinement, modification and concrete definition of experimental laws and theories. Considering the hypothesis as a form of “the development of natural science, as it thinks,” Engels put forward a provision on the relationship of hypotheses with laws and theories as forms of true knowledge.

The scientific hypothesis is always advanced in the context of the development of science to solve a specific problem with the aim of explaining new experimental data or eliminating the contradictions of the theory with negative experimental results. The replacement of the hypothesis in the development of science by another, more appropriate, does not mean acknowledgment of its falsity and uselessness at a certain stage of cognition: the nomination of a new hypothesis, as a rule, rests on the results of checking the old one (even if these results were negative). Therefore, the promotion of the hypothesis ultimately proves to be a necessary historical and logical stage in the development of another, a new hypothesis. For example, Planck’s development of the quantum hypothesis relied both on the conclusions obtained in the framework of the classical theory of radiation and on the negative results of testing his first hypothesis. The consideration of truth as a process taken together with the result leads to the conclusion that any relatively completed stage of cognition acting in the form of relative truths (experimental laws, theories) cannot be torn away from the process of one’s own becoming. The development of theories and the construction of applied models always requires the introduction of some auxiliary hypotheses that form a single whole with the original theory, mutually reinforcing each other and ensuring the progressive growth of scientific knowledge. So, in particular, the use of quantum mechanics as a theoretical basis for predicting the properties of various chemicals is impossible without the introduction of special hypotheses.

As scientific propositions, hypotheses must satisfy the condition of principled verifiability, meaning that they possess the properties of falsification (refutation) and verifiability (confirmation). However, the presence of such properties is necessary, but not sufficient condition for scientific hypotheses. The property of falsifiability fairly strictly fixes the alleged nature of scientific hypotheses. Restricting the universality of previous knowledge, and also revealing the conditions under which it is possible to preserve the partial universality of a statement about laws, the property of falsifiability provides a relatively discontinuous character of the development of scientific knowledge. The verifiability of the hypothesis makes it possible to establish and verify it with respect to empirical content. The greatest heuristic value is the confirmation of such facts and experimental laws, the existence of which could not be expected before the hypothesis being put forward. For example, the quantum hypothesis proposed by Einstein in 1905 after almost a decade was confirmed by Millikan’s experiments. The property of verifiability serves as an empirical basis for the processes of formation and development of a hypothesis and other forms of theoretical knowledge, stipulating a relatively uninterrupted character of the development of science. At the same time, the probabilistic or comparative evaluation of competing hypotheses in relation to a class of already established facts has methodological significance.

The heuristic role of the method of hypotheses in the development of scientific knowledge was reflected in hypothetical-deductive theories, which are deductively organized systems of hypotheses of various degrees of generality. Such theories are incomplete, which opens the possibility of their expansion and concretization due to additional hypotheses, applied models. All this ultimately ensures sufficient breadth and flexibility of the application of hypotheses and other developed forms of theoretical knowledge to reflect complex objects and processes of objective reality.