The discovery-justification distinction was taken for granted for a long time by old-timer philosophers of science. One of Kantorovich’s objections is that ‘discovery and justification are inseparable’. The paper explains the nature of the DJ distinction and why it seemed to the old-timers to be important and correct.
Kantorovich (1993, p. 97-99) argues that the context of discovery and justification is one of the main concepts of logical empiricism and it is one of weak points of traditional philosophy of science. According to the author, the context of discovery is about the actual method and process that has allowed a new idea to come forth while the context of justification is about the manner in which the new idea is evaluated. The author points out that the old school of Baconian inductive reasoning postulated laws and theories by applying inductive rules on observations. However, Kantorovich arrives at laws and theories by using the method of creative hypothesis. He has used the example of discoveries of Ampere who postulated the empirical laws of electricity. Kantorovich argues that by using inductive reasoning, the electric and magnetic phenomenon were deduced. However, to find the theory behind circulating currents, Ampere had to use his creative imagination. The creative process is not defined or covered by any rules and that the actual justification would be derived only after testing the predictions.
Kantorovich further argues that there are two approaches for justification of discoveries and they are generationism and consequentialism. In the first case, theories can be justified only if they can be derived from observation. An idea of theory is generationistically justified by showing that it can be derived from established premises and these can be observations, first principles or established theoretical foundations. So a hypothesis is justified if it was generated by derivation from established premises and if it can be derived by rational reconstruction of the process of discovery. On the other hand, the consequentialists argues that theories are tested only as per the success of their predictions. Therefore, a theory can be refuted based on the false predictions and this does not depend on the manner in which it was derived.
Kantorovich (1993, p. 99) points out that the modern logistic version of the D-J distinction was put forth by Hans Reichenbach who argued that the task of epistemology and philosophy of science is to deal only with the context of justification while the context of discovery refers to psychology and this can deal with the actual processes of thinking, However, the context of discovery is not excluded from the domain philosophy of science since some laws can be derived at by inductive reasoning. The epistemology is prescriptive because of the logical force while psychology would describe the actual manner in which the idea is obtained. Rather Reichenbach refers to justification by logical reconstruction rather than to empirical confirmation and this puts him as a generationist and not a consequentialist. In practical terms, justification is used with consequentialist interpretation. Logic is the key to a wider interpretation; the new observational result has to have some logical relation with the said hypothesis.
Kantorovich (1993, p. 99) argues that there are three theses that are built into the D-J distinction, there is a demarcation between two contexts; only the context of justification is subject to logical analysis and that the descriptive science such as psychology is not relevant to the context of justification. Therefore, the context of discovery can be dealt with by descriptive science while that of justification is prescriptive or normative. Practitioners of this view maintain that philosophy of science is a logic of science that should deal only with the context of justification. The most important consequences are that the information about the context of discovery is not related to justification. This argument follows that the context of discovery is logical and therefore it does not have any epistemic or justificatory force and the justification is concerned with the final result of discovery or a statement. Therefore, the manner in which the scientist deduced at the result does not matter. Therefore, the proponents of the D-J distinction a philosophy of science is a logic of science and a logic of science of justification or evaluation.
Kantorovich (1993, p. 101-102) has raised some objection to the D-J distinction and he argues that the two contexts are inseparable and that each context heavily depends on the elements of the other. The context of discovery is contaminated with justification of evaluation and evaluation is an internal part of the discovery process as an entity would be considered a discovery only if it was proved to be true. Therefore, while speaking of a theory, one would not say as it as being discovered unless it has been confirmed. When a scientist proposes a theory and did not prove it, he would not be considered as a discovery of the theory even if it was proved later. The act of discovery is about association and confirmation and these two acts may be simultaneous especially if the discovery is very important. However, the discovery is made when it can be seen that the idea helps to solve a problem. Therefore discovery is about selection and this is an evaluative act. This can best be shown by saying the Democritus first theorised the existence of atoms but the credit of discovery atoms goes to Dalton who actually proved their existence. Again while viewing the sky with a telescope, if one sees a new star, then the act of discovery and justification are one and the same. In generational discovery, the process of generation may precede the act of confirmation and here the theory is constructed, tested and then confirmed. Hence, the context of generation and the context of justification can be separated.
The paper has D-J context a rationale is thus found to be more tenable as it creates dependencies on both the context of discovery and justification. However, a simple example would show that both discovery and justification should adhere to the arguments of common logic. As an example, a cockerel crows every morning, just before sunrise and they have been doing this since times immemorial. There the postulation of a theory that suggests that the sunrise is wholly dependent on the cocks crowing and further justifying it with years of observation are wrong. Logic and intervention should prevail here and an attempt should be made to see if the sun still rises even when the cock has been taken away. Common logic tells us that there is not relation between the cocks crowing and sun rise. However, if the research into the phenomenon of deep space, sub atoms and other scientific observations that are rarefied come into the picture, it would be difficult to apply logic or create the D-J link.
References
Kantorovich Aharon. July 1993. Scientific Discovery: Logic and Tinkering. State University of New York Press.