1. Since psychology has for its object, not specific contents of experience, but general experience in its immediate character, it can make use of no methods except such as the empirical sciences in general employ for the determination, analysis, and causal synthesis of facts. The circumstance, that natural science abstracts from the subject, while psychology does not, can be no ground for modifications in the essential character of the methods employed in the two fields, though it does modify the way in which these methods are applied.
The natural sciences, which may serve as an example for psychology in this respect, since they were developed earlier, make use of two chief methods: experiment and observation. Experiment is observation connected with an intentional interference on the part of the observer, in the rise and course of the phenomena observed. Observation, in its proper sense, is the investigation of phenomena without such interference, just as they are naturally presented to the observer in the continuity of experience. Wherever experiment is possible, it is always used in the natural sciences; for under all circumstances, even when the phenomena in themselves present the conditions for sufficiently exact observation, it is an advantage to be able to control at will their rise and progress, or to isolate the various components of a composite phenomenon. Still, even in the natural sciences the two methods have been distinguished according to their spheres of application. It is held that the experimental methods are indispensable for certain problems, while in others the desired end may not infrequently be reached through mere observation. If we neglect a few exceptional cases due to special relations, these two classes of problems correspond to the general division of natural phenomena into processes and objects.
Experimental interference is required in the exact determination of the course, and in the analysis of the components, of any natural process such as, for example, light-waves or sound-waves, an electric discharge, the formation or disintegration of a chemical compound, and stimulation and metabolism in plants and animals. As a rule, such interference is desirable because exact observation is possible only when the observer can determine the moment at which the process shall commence. It is also indispensable in separating the various components of a complex phenomenon from one another.
As a rule, this is possible only through the addition or subtraction of certain conditions, or a quantitative variation of them.
The case is different with objects of nature. They are relatively constant; they do not have to be produced at a particular moment, but are always at the observer's disposal and ready for examination. Here, then, experimental investigation is generally necessary only when the production and modification of the objects are to be inquired into. In such a case, they are regarded either as products or components of natural processes and come under the head of processes rather than objects. When, on the contrary, the only question is the actual nature of these objects, without reference to their origin or modification, mere observation is generally enough. Thus, mineralogy, botany, zoology, anatomy, and geography, are pure sciences of observation so long as they are kept free from the physical, chemical, and physiological problems that are, indeed, frequently brought into them, but have to do with processes of nature, not with the objects in themselves.
2. If we apply these considerations to psychology, it is obvious at once, from the very nature of its subject-matter, that exact observation is here possible only in the form of experimental observation; and that psychology can never be a pure science of observation. The contents of this science are exclusively processes, not permanent objects. In order to investigate with exactness the rise and progress of these processes, their composition out of various components, and the interrelations of these components, we must be able first of all to bring about their beginning at will, and purposely to vary the conditions of the same. This is possible here, as in all cases, only through experiment, not through pure introspection. Besides this general reason there is another, peculiar to psychology, that does not apply at all to natural phenomena.