An Oscillatory Model of the Growth of Scientific Knowledge

Abstract

The purpose of this paper is to propose a new model of scientific progress which is based on an analogy with the well-known oscillatory reactions which occur in chemical systems. In these reactions the concentrations of reactive intermediates show a rhythmic increase and decrease which can be experimentally followed by observing colour change, variation of pH or electric conductivity. In the more usual type of chemical reaction the concentration of the reactive intermediates is approximately constant and the reactants are smoothly converted into the products in a manner analogous to a Popperian 'continuous revolution' of scientific progress. In the analogy which I will be developing the reactants are equivalent to the 'protocol sentences' described in section 26 of the Logic of Scientific Discovery (Popper [1959]) (i.e. the unorganized sense data with which the scientist is continuously bombarded every second of the day); the products are equivalent to the 'basic statements' described in section 28 of Popper [1959] (i.e. the accepted scientific facts which have emerged like tempered steel from the furnace of crucial experiments) and the reactive intermediates are equivalent to the plurality of scientific hypotheses which are alternately generated then refuted as proposed by Popper in 1956 in the 'Postscript' but subsequently only published in Popper [1983]. A Kuhnian model of scientific progress with respect to time t would be represented by Diagram 1, an old fashioned Popperian model of scientific progress by Diagram 2, and a modern Popperian model by Diagram 3. An essential feature of the latter model is the dynamic interaction of hypotheses with protocol sentences which leads to the production of basic statements-a proliferation of hypotheses is necessary for the new experiments which will generate the new scientific facts; then a reduction in the number of hypotheses is necessary for the next period of advance. In the field of Chemistry a large number of molecules for which we possess the technology to synthesize, however, remain uninvestigated because we do not consider them to be sufficiently interesting to be synthesized at the present time. For example, the molecule 1-bromobicyclo(2.2.2.)-octane could have been synthesized around the year 1900 and its peculiar resistance to