ABSTRACT
I N this book we shall be concerned with the study of scientific procedure. I t is made up of the study of scientific inference and the study of scientific practice. In addition to inference, certain other general features of method play an indispensable part in science. Indeed it cannot be overstressed that scientific inference and scientific method are interwoven and inseparable. The subject is commonly described as "methodology" or as "the study of scientific method". Either description is, however, at once too wide and too narrow. On the one hand either would allow us to include minute details concerning the techniques used in making experiments-but the methodologist is not interested in whether it is better to seal a joint in a piece of glass-tubing by means of plasticine or rubber. On the other hand the word "method" draws the stress away from processes of inference, which play an enormous part in science. We may, however, conveniently reserve "methodology" or "scientific method" for the study of the general features of scientific practice in which the role of inference is not stressed. In this book scientific inference is much more extensively discussed than methodology. It will be convenient, therefore, to introduce the word "metascience" as a name for the study of scientific inference. It should be observed, moreover, that in this book metascience is being restricted to scientific procedure in natural science; mathematics is a science in a different sense; the corresponding metascientific study of it is deductive logic; but this study will not be included.
