Essentials of Scientific Method

A. Wolf

• Publisher: Routledge
• Year: 2019
• ISBN:
  9780429620454
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Originally published in 1925, when it was published, this book was intended to give an up to date, concise account of the aim and methods of science with regards to Psychology. It contains chapters on various scientific methods such as the Evolutionary or Genetic Method, the Method of Difference and The Method of Residues, and chapters on probability and the laws of nature.

CHAPTER V

THE SIMPLER INDUCTIVE METHODS

§ 1. Classification and Law.

The scientific search for general truths is satisfied in some measure by the discovery of natural classes through the Method of Classification, and by ascertaining evolutionary sequences with the aid of the Comparative Method. In the former case, we discover certain uniformities of co-existence among the groups of essential characteristics of the several natural classes. In the latter case, we discover certain uniformities of sequence among various complex phenomena, which follow one another as successive phases or stages of an evolutionary process. On the strength of our knowledge of a uniformity of co-existence among the attributes of a class, it is possible to infer from the presence of some class-characteristics in an object (e.g. the frontal horns and the hoofs of a quadruped) also the presence of certain other characteristics (e.g. the possession of a ruminant stomach, and graminivorous habits¹); and on the strength of our knowledge of a uniformity of sequence among the stages of the evolution of certain phenomena, one might anticipate the coming of a subsequent stage from the observation of an earlier one, or imaginatively interpolate something between two known stages.

But such uniformities as the foregoing are for the most part only empirical, and not altogether satisfactory. Science looks as far as possible for what can be more or less adequately proved. Even the relative values of different classifications will be assessed according to their usefulness for real inductions. Now all the proofs of general truths take one of two forms. One of them is the type with which we are familiar from geometry, where it can be shown by sheer intuition, or by deductive reasoning from intuition, that certain attributes must be present where certain other attributes are present; but the fundamental uniformities of natural science cannot be established in that way. The other method of proving uniformities is by ascertaining either the direct or indirect causal connection between the terms of the uniform relationship. In both forms of proof we try to establish relations between conditions and consequents; only in Mathematics (also in Logic) we are concerned with rational conditions (or reasons) and consequents, while in natural science we are concerned with physical conditions and results (in psychology with psychical conditions and results which are intimately connected with physical or physiological conditions and results).

The world, as we see it, is a vast complex of incessantly changing things, which the human mind endeavours to grasp by mentally, and sometimes also physically, analysing into simpler constituents, and ascertaining the laws or regularities of their connections, or their correlations, if there be such. The facts themselves do not manifest their intimate relations with one another. We can only solve the riddle, if at all, by surmising what the relations may be. Such surmises are only fruitful when they have been preceded by close observation of the facts and are followed up by a still more searching observation and (where possible) experimentation.

§ 2. The Five Canons or Methods of Induction.

The kinds of observations by which the man of science is led to surmise a real connection between certain facts, and the kinds of observation by which he then proceeds to test his surmise, or hypothesis, are often very similar. Their general character has been formulated in the five so-called Canons of Induction. These are not the only methods of ascertaining laws, or uniformities, or regularities among the phenomena of Nature. We have already described two other methods of doing so, and yet other methods will be explained in due course. But the five canons or methods of induction are important all the same.

The principle underlying these canons is this. If, other things remaining essentially the same, a certain factor or circumstance cannot be omitted, or quantitatively changed, without changing a certain phenomenon, then that factor, or circumstance, is a condition of that phenomenon, or, in other words, is intimately connected with it. Assuming, as we generally do, that things and events are not merely a matter of chance, but are the results of operative conditions, we examine instances of the phenomenon, in which we are interested, under sufficiently varied circumstances to enable us to detect what it is that cannot be removed or altered without removing or altering the phenomenon in question. Not that the presence of an element of chance in the universe is to be ruled out ab initio. We shall return to this point in due course. But it is order and regularity that have helped man most in his struggle for existence. It is order also that satisfies most the rational tendency and the aesthetic sense of man. So we naturally look for order first, and only reluctantly relinquish our search for it when we are baffled in our quest.

Now the process of ascertaining what is indispensable to a certain phenomenon may assume one of two forms, a direct form and an indirect one. In the direct form it is shown by observation or experiment that (a) the elimination, or (b) the quantitative variation of a certain factor or antecedent is followed by (a) the elimination, or (b) the quantitative variation of the phenomenon under investigation, although all other relevant factors have remained the same. In the indirect form of the inductive process it is shown that, so long as a certain antecedent remains operative, no change in any of the other relevant circumstances makes any material difference to the phenomenon under investigation, which must, therefore, be intimately connected with the constant antecedent. The first type of the direct form is known as the Method of Difference, the second type of the same form is called the Method of Concomitant Variations. The indirect form is known as the Method of Agreement. Of the two remaining Canons or Methods, one is known as the Method of Residues, and is really a slight modification of the Method of Difference, while the other is known as the Joint Method of Agreement and Difference (also as the Method of Exclusion, or as the Double Method of Agreement), and is a kind of approximation to the Method of Difference, secured by supplementing the Method of Agreement in certain directions.

The several inductive methods have different degrees of cogency. The Methods of Difference and of Concomitant Variations are the most conclusive. But even these methods cannot always be applied rigorously. When, as sometimes happens, the phenomena under investigation are not sufficiently under the control of the investigator, he may not be able to secure the precise kinds of instances required for the strict application of these methods. But, in science, as in life generally, if one cannot command the best means, one tries the next best, and so on. In such cases, one usually...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Original Title Page
• Original Copyright Page
• Preface
• Table of Contents
• I. Introduction
• II. The Chief Cognitive Processes in the Service of Science
• III. Classification and Description
• IV. The Evolutionary and Comparative Methods
• V. The Simpler Inductive Methods
• VI. The Statistical Method
• VII. The Deductive-Inductive Method
• VIII. Order in Nature and Laws of Nature
• IX. Scientific Explanation
• X. Probability
• Select List of Books
• Index